- Email: [email protected]
Serial blood lactate measurements and its prognostic significance in intensive care unit management of aneurysmal subarachnoid hemorrhage patients
Journal of Critical Care 41 (2017) 229–233
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Serial blood lactate measurements and its prognostic significance in intensive care unit management of aneurysmal subarachnoid hemorrhage patients Tomoya Okazaki a,1, Toru Hifumi a,⁎,1, Kenya Kawakita a, Hajime Shishido a, Daisuke Ogawa b, Masanobu Okauchi b, Atsushi Shindo b, Masahiko Kawanishi b, Shigeaki Inoue c, Takashi Tamiya b, Yasuhiro Kuroda a a b c
Emergency Medical Center, Kagawa University Hospital, 1750-1 Ikenobe, Miki, Kita, Kagawa 761-0793, Japan Department of Neurosurgery, Kagawa University Hospital, 1750-1 Ikenobe, Miki, Kita, Kagawa 761-0793, Japan Department of Emergency and Critical Care Medicine, Tokai University Hachioji Hospital, Ishikawa-cho 1838, Hachioji City, Tokyo 192-0032, Japan
a r t i c l e
i n f o
Keywords: Aneurysmal subarachnoid hemorrhage Lactate Neurological outcome
a b s t r a c t Purpose: This study assesses the behavior of serial blood lactate measurements during intensive care unit (ICU) stay to identify prognostic factors of unfavorable neurological outcomes (UO) in patients with aneurysmal subarachnoid hemorrhage (SAH). Methods: We retrospectively reviewed all patients who were consecutively hospitalized with SAH between 2009 and 2016. Arterial blood lactate levels were routinely obtained on admission and every 6 h in the ICU. Univariate/ multivariate analyses were performed to identify independent predictors of UO (modified Rankin scale of 3–6 upon hospital discharge). Results: There were 145 patients with 46% of UO. Initially, increased lactate levels reached maximum levels during the first 24 h and then decreased to within the normal range. Then, the levels slightly increased again to within the normal range for the next 24 h, especially in UO. On multiple regression analysis, lactate levels measured at 24 h, and 48 h after admission were strong predictors of UO. Lactate level measured at 48 h after admission demonstrated the greatest accuracy and the highest specificity (area under the curve, 0.716; sensitivity, 40%; specificity, 92.1%). Conclusions: The lactate level at 48 h after admission was the most accurate predictor of UO with a high specificity in SAH patients. © 2017 Elsevier Inc. All rights reserved.
1. Introduction Aneurysmal subarachnoid hemorrhage (SAH) requires management in the intensive care unit (ICU) for both prevention and early detection of subsequent critical complications, as well as for appropriate intervention [1-3]. Serial blood lactate measurements during ICU management
Abbreviations: SAH, aneurysmal subarachnoid hemorrhage; ICU, intensive care unit; IRB, institutional review board; ROC, receiver operating characteristic curve; AUC, area under the curve. ⁎ Corresponding author at: 1750-1 Ikenobe, Miki, Kita, Kagawa 761-0793, Japan. E-mail addresses: [email protected] (T. Hifumi), [email protected] (K. Kawakita), [email protected] (H. Shishido), [email protected] (D. Ogawa), [email protected] (M. Okauchi), [email protected] (A. Shindo), [email protected] (M. Kawanishi), [email protected] (S. Inoue), [email protected] (T. Tamiya), [email protected] (Y. Kuroda). 1 Drs. Okazaki and Hifumi contributed equally to this work.
http://dx.doi.org/10.1016/j.jcrc.2017.06.001 0883-9441/© 2017 Elsevier Inc. All rights reserved.
have been reported to be clinically more reliable and accurate than an absolute lactate value as a marker of global tissue hypoxia in various critical conditions [4-6]. Although studies assessing associations between blood lactate level and medical complications in SAH are limited, a study by Satoh et al. reported that increased blood lactate level on admission was correlated with the early onset of neurogenic pulmonary edema in patients with SAH [7]. In addition, the maximal lactate level within 24 h after admission and lactate levels on admission were recently reported to be prognostic factors of unfavorable neurological outcomes in patients with SAH [8,9]; however, serial measurements of lactate levels during ICU stay have not been evaluated in patients with SAH. The purpose of this study was to assess the behavior of serial blood lactate measurements during ICU stay and to identify the prognostic factors of unfavorable neurological outcomes in patients with SAH in the acute phase.
230
T. Okazaki et al. / Journal of Critical Care 41 (2017) 229–233
2. Materials and methods 2.1. Study design and setting Kagawa University Hospital is a 613-bed academic teaching institution with an 8-bed neurointensivist-managed ICU [10]. The protocol of this single-center, retrospective, case-control study, which was performed by retrospective review of medical records, was approved by the institutional review board (IRB) of Kagawa University Hospital and was conducted in accordance with the ethical standards established in the 1964 Declaration of Helsinki and its later amendments. The IRB waived the requirement for patient consent due to the retrospective nature of the study.
Categorical comparisons were conducted using the Fisher's exact test. Univariate and multivariate regression analyses (with step-wise variable selection) were conducted to identify independent factors predictive of unfavorable neurologic outcomes in the study population. Covariates with age, sex (male), Hunt & Kosnik grade, Fisher group, treatment modality (coil or clip), and lactate levels at each time were included in the multivariate analysis. Receiver operating characteristic (ROC) curve and the area under the curve (AUC) were estimated based on the Mann–Whitney U test. For each continuous variable, the cut-off value with the best combination of sensitivity and specificity was identified. Statistical analyses were performed using JMP® statistical software (version 11; SAS Institute, Cary, NC, USA). A twosided p value of b 0.05 was considered statistically significant for all analyses.
2.2. Study participants and inclusion criteria All patients ≥18 years old who were consecutively hospitalized for a confirmed diagnosis of SAH and had at least five arterial lactate measurements between January 1, 2009 and June 30, 2016 were retrospectively enrolled. Patients who received only comfort care within 24 h of admission were excluded. Patients with unknown onset of SAH were also excluded. 2.3. General management of SAH in the ICU Management was conducted in line with the Guidelines for the Management of Aneurysmal Subarachnoid Hemorrhage of the American Heart Association/American Stroke Association [11]. In addition to general intensive care, all patients were monitored for clinical deterioration or cerebral infarction arising due to delayed cerebral ischemia. Fluid management was targeted at maintaining euvolemia. Prophylactic hemodynamic support, hemodilution, and administration of mannitol or hypertonic saline solution were not performed [10,12]. 2.4. Measurement of lactate levels Arterial blood gas analyses, including lactate levels, were routinely performed on admission and every 6 h in the ICU. Additional measurements were carried out as needed by critical care physicians. Arterial blood gas measurements from blood samples collected in heparinized blood gas syringes were measured at 37 °C using a blood gas analyzer [12]. 2.5. Data sampling The following data were collected: age, sex, Hunt and Kosnik grade, Fisher group, treatment modality (coil or clip), serial arterial lactate, modified Rankin scale (mRS) upon hospital discharge, length of ICU stay, length of hospital stay, and hospital mortality. 2.6. Outcome measures The primary outcome was the association of serial lactate levels measured in the ICU with an unfavorable neurologic outcome, which was assessed by the mRS upon hospital discharge. The neurologic outcome was defined as unfavorable when the mRS score was 3–6 and as favorable when the mRS score was 0–2. The secondary outcome was to determine the most accurate predictor in serial lactate measurements. 2.7. Statistical analysis Demographic factors and baseline characteristics were summarized using descriptive statistics. The groups were compared using the Student's t-test or Mann–Whitney U test, as deemed appropriate.
3. Results 3.1. Baseline characteristics of the study population and comparison of baseline characteristics between unfavorable and favorable outcomes The study population included 145 patients (mean age, 62.5 years; 44 men) (Table 1). Neurological outcomes were unfavorable in 46.2% of these patients. Patients with unfavorable neurological outcomes were significantly older than those with favorable outcomes [mean ± standard deviation, 69.5 ± 15.0 vs. 56.5 ± 15.0 years, respectively, p b 0.01]. ICU stay and hospital stay were significantly longer in the unfavorable neurological outcome group than the favorable outcome group [median, interquartile range; 17 (13 − 21) vs. 14 (12–16) days, p b 0.01; 46 (26–75) vs. 24 (21 − 31) days, p b 0.01, respectively].
Table 1 Characteristics of the study population. Variables
All patients (n = 145)
Unfavorable neurological outcomes group (n = 67)
Favorable neurological outcomes group (n = 78)
p value
Age (years) Gender (male) H&K grades I II III IV V Fisher grades 1 2 3 4 Treatment modality Clip Coil Modified Rankin Scale 0 1 2 3 4 5 6 ICU stay Hospital stay Hospital mortality
62.5 ± 16.3 44 (30.3)
69.5 ± 15.0 19 (28.4)
56.5 ± 15.0 25 (32.1)
b0.01 0.63 b0.01
11 (7.6) 55 (37.9) 37 (25.5) 28 (19.3) 14 (9.7)
2 (3.0) 12 (17.9) 21 (31.3) 22 (32.8) 10 (14.9)
9 (11.5) 43 (55.1) 16 (20.5) 6 (7.7) 4 (5.1)
1 (0.7) 34 (23.4) 99 (68.3) 11 (7.6)
0 (0.0) 8 (11.9) 54 (80.6) 5 (7.5)
1 (1.3) 26 (33.3) 45 (57.7) 6 (7.7)
36 (24.8) 109 (75.2)
14 (20.9) 53 (79.1)
22 (28.2) 56 (71.8)
29 (20.0) 22 (15.2) 27 (18.6) 22 (15.2) 24 (16.6) 12 (8.3) 9 (6.2) 15 [12–19] 28 [22–53] 9 (6.2)
0 (0.0) 0 (0.0) 0 (0.0) 22 (32.8) 14 (35.8) 12 (17.9) 9 (13.4) 17 [13−21] 46 [26–75] 9 (13.4)
29 (37.2) 22 (28.2) 27 (34.6) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 14 [12–16] 24 [21−31] 0 (0.0)
b0.01
0.31
b0.01
b0.01 b0.01 b0.01
Data are expressed as mean ± standard deviation, number (percentage), or median [interquartile range]. H&K, Hunt and Kosnik.
T. Okazaki et al. / Journal of Critical Care 41 (2017) 229–233
231
Fig. 1. Serial measurements of lactate in all study patients. The boxes are the 25th to 75th percentiles and the whiskers are the 5th to 95th percentiles.
Serial lactate measurements in all study patients are shown in Fig. 1. Overall, initially increased lactate levels decreased to normal ranges during the first 24 h. 3.2. Alteration of arterial lactate levels between the unfavorable and favorable neurological outcome groups during ICU stay Serial lactate measurements in the unfavorable and favorable neurological outcome groups are shown in Fig. 2. The serum lactate level at each time point was greater in the unfavorable neurological outcome group than the favorable neurological outcome group and initial lactate levels increased to the maximum levels during the first 24 h and then decreased to within normal ranges. Then, the lactate levels slightly increased again to within the normal range for the next 24 h, especially in patients with unfavorable neurological outcome.
3.3. Independent predictors of unfavorable neurologic outcome The results of multivariate logistic regression models are shown in Table 2. Lactate levels measured at 24 h [odds ratio (OR), 2.42; 95% confidence interval (CI), 1.38–4.75; p b 0.01], 36 h (OR, 3.71; 95% CI, 1.80– 8.70; p b 0.01), and 48 h after admission (OR, 3.39; 95% CI, 1.26– 10.81; p = 0.01) were strong predictors of unfavorable neurological outcomes (Supplemental Table 1 demonstrates the result of multivariate logistic regression models).
3.4. ROC curve analysis ROC curves of unfavorable neurological outcomes based on serial lactate levels at each time point were constructed. The respective
Fig. 2. Serial measurements of lactate in the unfavorable and favorable neurological outcome groups. UO, unfavorable neurological outcomes; FO, favorable neurological outcomes. Patients in the unfavorable outcome group are indicated in gray and those in the favorable outcome group are indicated in white. The boxes are the 25th to 75th percentiles and the whiskers are the 5th to 95th percentiles.
232
T. Okazaki et al. / Journal of Critical Care 41 (2017) 229–233
Table 2 Multivariate logistic regression models. Model
Variables
Adjusted odds ratio
95% confidence interval
p value
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
Lactate on admission Lactate at 12 h after admission Lactate at 24 h after admission Lactate at 36 h after admission Lactate at 48 h after admission Lactate at 60 h after admission Lactate at 72 h after admission
1.38 1.33 2.45 3.71 3.41 4.21 1.07
1.02–1.93 0.91–2.00 1.39–4.84 1.80–8.69 1.26–10.93 1.00–21.16 0.34–3.59
b0.05 0.14 b0.01 b0.01 0.01 0.05 0.91
Each model was adjusted according to age, sex (male), Hunt & Kosnik grade, Fisher grade and treatment modality (coil or clip). Age and Hunt & Kosnik grade were significant factors in all models.
AUCs, cut-off levels, sensitivities, and specificities for prediction of unfavorable neurologic outcomes are shown in Table 3. The lactate level measured at 48 h after admission was the most accurate predictor with greater specificity than at other time points (AUC, 0.716; cut-off level, 1.1 mmol/L; sensitivity, 40%; specificity, 92.1%). 3.5. Details of 6 cases with favorable neurological outcome in spite of lactate level N 1.1 mmol/L at 48 h after admission The characteristics of these patients are shown in Supplemental Table 2. In four patients (nos. 1, 2, 4, and 5), extubation was completed by 48 h after admission and these patients complained of severe headache, which was not resolved with analgesics at 48 h. Two patients (nos. 3 and 6) developed serious complications. Patient no. 3 developed both stress-induced cardiomyopathy and neurogenic pulmonary edema, and patient no. 6 was in shock status due to rupture of a pancreaticoduodenal aneurysm. 3.6. Comparison of baseline characteristics between patients with lactate levels at 48 h N 1.1 mmol/L and patients with lactate levels at 48 h ≤ 1.1 mmol/L in the favorable and unfavorable outcome groups The portion of male patients with lactate levels at 48 h ≤ 1.1 mmol/L was significantly higher than that of patients with lactate levels at 48 h N 1.1 mmol/L in the unfavorable outcome group (41% vs. 11.5%, p = 0.01). The same tendency was observed in the favorable outcome group (34.3% vs. 16.7%, p = 0.38) (Supplemental Table 3 demonstrates the result of comparison of baseline characteristics between patients with lactate levels at 48 h N 1.1 mmol/L and patients with lactate levels at 48 h ≤ 1.1 mmol/L in the favorable and unfavorable outcome groups). 3.7. The association between timing of treatment (coil or clip) and alteration of lactate levels during the first 48 h Initial treatment (coil or clip) completed within 30 h after admission in both 95% (138/145) of all study patients and 94% (63/67) of patients with unfavorable neurological outcome. (Supplemental Table 4 demonstrates the result of comparison of timing of initial treatment between the unfavorable neurological outcome group and favorable neurological outcome group in the coil and clip groups).
4. Discussion The main findings of the current study were as follows: (1) in the unfavorable outcome group, initially increased lactate levels decreased to within the normal range for the first 24 h. Then, lactate levels slightly increased again to within the normal range for the next 24 h. (2) The lactate level at 48 h after admission was the most accurate predictor of unfavorable neurological outcomes in the acute phase (from on admission to 72 h after admission) with a specificity N90% in patients with SAH. Serum lactate levels during ICU management of SAH patients can be influenced by multifactorial factors. Not only volume status, but also aerobic glycolysis caused by excessive catecholamine levels and impaired lactate clearance reflecting hepatic dysfunction due to sedatives and antiepileptic drugs [13,14], may increase lactate levels. However, when limited to the first 48 h, an excess in catecholamine is considered the main reason for the increase of lactate levels because hypovolemic shock and hepatic dysfunction were rarely seen in our study populations (data not shown). Ogura et al. examined catecholamine levels during the first 24 h after SAH and observed high catecholamine levels in samples obtained within the first 6 h, but levels in samples obtained around 24 h were mostly normal [15]. Benedict et al. reported that catecholamine levels after neurosurgery for SAH were altered and initial high catecholamine levels were normalized at 12 h (estimated to be approximately 24 h after admission), but were increased again around 24 h after neurosurgery (estimated to be approximately 36 h after admission) [16]. Our current data of alteration of lactate levels in the unfavorable outcome group were paralleled with the above-mentioned alteration of catecholamine levels (Fig. 2). That is, initially increased lactate levels reached maximum levels during the first 24 h and then decreased to within the normal range. Interestingly, lactate levels slightly increased again to within the normal range for the next 24 h. Thus, a distinct difference was observed between the favorable and unfavorable neurological outcome groups at 48 h after admission. Therefore, we believe that the lactate level at 48 h is the most accurate predictor of neurological outcome, as compared to two time points (on admission and maximum level during the first 24 h) as previously reported. van Donkelaar et al. reported an association between the maximal lactate level within 24 h after admission and unfavorable neurological outcomes in patients with SAH [8]. Aisiku et al. reported that elevated
Table 3 Receiver operating characteristic curve analysis of serial measurements of arterial blood lactate levels. Variables
AUC
SE
Cut-off level
Sensitivity
Specificity
Lactate on admission Lactate at 12 h after admission Lactate at 24 h after admission Lactate at 36 h after admission Lactate at 48 h after admission Lactate at 60 h after admission Lactate at 72 h after admission
0.628 0.623 0.676 0.712 0.716 0.695 0.629
0.127 0.160 0.311 0.362 0.540 0.613 0.463
2.9 mmol/L 1.4 mmol/L 1.0 mmol/L 1.0 mmol/L 1.1 mmol/L 0.8 mmol/L 0.9 mmol/L
41.3 55.2 61.2 59.7 40.0 60.9 48.4
82.1 68.0 70.5 77.9 92.1 68.0 72.6
AUC, area under the curve; SE, standard error.
T. Okazaki et al. / Journal of Critical Care 41 (2017) 229–233
serum lactate levels on admission may be predictive of mortality [9]. In the current study, re-evaluation of serial lactate measurements showed that lactate levels at 48 h after admission provided the highest AUC, in accordance with previously reported predictors [8,9]. This may be considered due to the nature of the prognostic factor because, in general, later points can predict more accurate prognosis. However, novel specificity for predicting unfavorable neurological outcome and analysis based on serial measurements of lactate during the first 72 h could be a strength. According to our current data, arterial lactate level measured at 48 h after admission N 1.1 mmol/L is a predictor of unfavorable neurological outcome with a specificity of 92.1%. Lactate measurements are quick, cheap, and more commonly used than other previously reported prognostic markers [17], such as S100β protein [18,19] and neuron-specific enolase [20,21]; therefore, lactate evaluation in itself offers a great advantage. On the other hand, six of 32 cases with lactate level N 1.1 mmol/L at 48 h achieved favorable neurological outcomes with intensive care. Detailed examinations of those cases revealed that patients with complications, such as stress-induced cardiomyopathy and hypovolemia, and severe headache, indicate that a higher catecholamine level may beyond be the cut-off value; therefore, caution is needed when applying these results clinically. According to the comparison of baseline characteristics between patients with lactate levels at 48 h N 1.1 mmol/L vs. ≤ 1.1 mmol/L in the favorable and unfavorable outcome groups, lactate clearance was more dominant in males than in females in both groups. Although the mechanism remains unknown, this tendency should be also kept in mind. We also examined the association between timing of initial treatment and alteration of lactate levels during the first 48 h, especially slightly re-increased lactate levels around 36–42 h in patients with unfavorable neurological outcome. Because 94% of patients with unfavorable neurological outcome completed initial treatment within 30 h, we do not believe that the timing of treatment affected the alteration of lactate levels. This study had several limitations. First, it was a retrospective cohort study conducted at a single center, which introduces potential selection bias. Moreover, uncontrolled confounding factors may exist. Second, it was not possible to follow-up neurological outcomes of patients after discharge. Third, we could not definitively exclude the possibility that changes in the patient population had an impact on neurological outcomes; for example, comorbidities were not examined in the current study. Fourth, recognition of elevation in lactate levels led to the adjustment of fluids and workup to reveal the etiology by critical care physicians; however, that approach was not protocol-based. Finally, the sample size in this study was relatively small. 5. Conclusions Initially increased lactate levels reached maximum levels during the first 24 h and then decreased to within the normal range. Then, lactate levels slightly increased again to within the normal range for the next 24 h in the unfavorable outcome group. The lactate level at 48 h after admission was the most accurate predictor of unfavorable neurological outcomes with a specificity N90% in SAH patients. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jcrc.2017.06.001. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
233
Conflict of interest The authors declare that they have no conflict of interest.
Acknowledgments We are grateful to all physicians and nurses at the study site for their crucial contribution to the successful completion of this study. References [1] Samuels O, Webb A, Culler S, Martin K, Barrow D. Impact of a dedicated neurocritical care team in treating patients with aneurysmal subarachnoid hemorrhage. Neurocrit Care 2011;14:334–40. [2] Josephson SA, Douglas VC, Lawton MT, English JD, Smith WS, Ko NU. Improvement in intensive care unit outcomes in patients with subarachnoid hemorrhage after initiation of neurointensivist co-management. J Neurosurg 2010;112:626–30. [3] Knopf L, Staff I, Gomes J, McCullough L. Impact of a neurointensivist on outcomes in critically ill stroke patients. Neurocrit Care 2012;16:63–71. [4] Bonizzoli M, Lazzeri C, Cianchi G, Boddi M, Cozzolino M, Di Valvasone S, et al. Serial lactate measurements as a prognostic tool in venovenous extracorporeal membrane oxygenation support. Ann Thorac Surg 2016. [5] Bakker J, Gris P, Coffernils M, Kahn RJ, Vincent JL. Serial blood lactate levels can predict the development of multiple organ failure following septic shock. Am J Surg 1996;171:221–6. [6] Vincent JL, Dufaye P, Berre J, Leeman M, Degaute JP, Kahn RJ. Serial lactate determinations during circulatory shock. Crit Care Med 1983;11:449–51. [7] Satoh E, Tagami T, Watanabe A, Matsumoto G, Suzuki G, Onda H, et al. Association between serum lactate levels and early neurogenic pulmonary edema after nontraumatic subarachnoid hemorrhage. J Nippon Med Sch 2014;81:305–12. [8] van Donkelaar CE, Dijkland SA, van den Bergh WM, Bakker J, Dippel DW, Nijsten MW, et al. Early circulating lactate and glucose levels after aneurysmal subarachnoid hemorrhage correlate with poor outcome and delayed cerebral ischemia: a twocenter cohort study. Crit Care Med 2016. [9] Aisiku IP, Chen PR, Truong H, Monsivais DR, Edlow J. Admission serum lactate predicts mortality in aneurysmal subarachnoid hemorrhage. Am J Emerg Med 2016; 34:708–12. [10] Egawa S, Hifumi T, Kawakita K, Okauchi M, Shindo A, Kawanishi M, et al. Impact of neurointensivist-managed intensive care unit implementation on patient outcomes after aneurysmal subarachnoid hemorrhage. J Crit Care 2016;32:52–5. [11] Connolly Jr ES, Rabinstein AA, Carhuapoma JR, Derdeyn CP, Dion J, Higashida RT, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/ American Stroke Association. Stroke 2012;43:1711–37. [12] Okazaki T, Hifumi T, Kawakita K, Shishido H, Ogawa D, Okauchi M, et al. Blood glucose variability: a strong independent predictor of neurological outcomes in aneurysmal subarachnoid hemorrhage. J Intensive Care Med 2016. [13] Bakker J, Nijsten MW, Jansen TC. Clinical use of lactate monitoring in critically ill patients. Ann Intensive Care 2013;3:12. [14] Fuller BM, Dellinger RP. Lactate as a hemodynamic marker in the critically ill. Curr Opin Crit Care 2012;18:267–72. [15] Ogura T, Satoh A, Ooigawa H, Sugiyama T, Takeda R, Fushihara G, et al. Characteristics and prognostic value of acute catecholamine surge in patients with aneurysmal subarachnoid hemorrhage. Neurol Res 2012;34:484–90. [16] Benedict CR, Loach AB. Sympathetic nervous system activity in patients with subarachnoid hemorrhage. Stroke 1978;9:237–44. [17] Hong CM, Tosun C, Kurland DB, Gerzanich V, Schreibman D, Simard JM. Biomarkers as outcome predictors in subarachnoid hemorrhage—a systematic review. Biomarkers 2014;19:95–108. [18] Weiss N, Sanchez-Pena P, Roche S, Beaudeux JL, Colonne C, Coriat P, et al. Prognosis value of plasma S100B protein levels after subarachnoid aneurysmal hemorrhage. Anesthesiology 2006;104:658–66. [19] Sanchez-Pena P, Pereira AR, Sourour NA, Biondi A, Lejean L, Colonne C, et al. S100B as an additional prognostic marker in subarachnoid aneurysmal hemorrhage. Crit Care Med 2008;36:2267–73. [20] Moritz S, Warnat J, Bele S, Graf BM, Woertgen C. The prognostic value of NSE and S100B from serum and cerebrospinal fluid in patients with spontaneous subarachnoid hemorrhage. J Neurosurg Anesthesiol 2010;22:21–31. [21] Mabe H, Suzuki S, Mase M, Umemura A, Nagai H. Serum neuron-specific enolase levels after subarachnoid hemorrhage. Surg Neurol 1991;36:170–4.
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Serial blood lactate measurements and its prognostic significance in intensive care unit management of aneurysmal subarachnoid hemorrhage patients Tomoya Okazaki a,1, Toru Hifumi a,⁎,1, Kenya Kawakita a, Hajime Shishido a, Daisuke Ogawa b, Masanobu Okauchi b, Atsushi Shindo b, Masahiko Kawanishi b, Shigeaki Inoue c, Takashi Tamiya b, Yasuhiro Kuroda a a b c
Emergency Medical Center, Kagawa University Hospital, 1750-1 Ikenobe, Miki, Kita, Kagawa 761-0793, Japan Department of Neurosurgery, Kagawa University Hospital, 1750-1 Ikenobe, Miki, Kita, Kagawa 761-0793, Japan Department of Emergency and Critical Care Medicine, Tokai University Hachioji Hospital, Ishikawa-cho 1838, Hachioji City, Tokyo 192-0032, Japan
a r t i c l e
i n f o
Keywords: Aneurysmal subarachnoid hemorrhage Lactate Neurological outcome
a b s t r a c t Purpose: This study assesses the behavior of serial blood lactate measurements during intensive care unit (ICU) stay to identify prognostic factors of unfavorable neurological outcomes (UO) in patients with aneurysmal subarachnoid hemorrhage (SAH). Methods: We retrospectively reviewed all patients who were consecutively hospitalized with SAH between 2009 and 2016. Arterial blood lactate levels were routinely obtained on admission and every 6 h in the ICU. Univariate/ multivariate analyses were performed to identify independent predictors of UO (modified Rankin scale of 3–6 upon hospital discharge). Results: There were 145 patients with 46% of UO. Initially, increased lactate levels reached maximum levels during the first 24 h and then decreased to within the normal range. Then, the levels slightly increased again to within the normal range for the next 24 h, especially in UO. On multiple regression analysis, lactate levels measured at 24 h, and 48 h after admission were strong predictors of UO. Lactate level measured at 48 h after admission demonstrated the greatest accuracy and the highest specificity (area under the curve, 0.716; sensitivity, 40%; specificity, 92.1%). Conclusions: The lactate level at 48 h after admission was the most accurate predictor of UO with a high specificity in SAH patients. © 2017 Elsevier Inc. All rights reserved.
1. Introduction Aneurysmal subarachnoid hemorrhage (SAH) requires management in the intensive care unit (ICU) for both prevention and early detection of subsequent critical complications, as well as for appropriate intervention [1-3]. Serial blood lactate measurements during ICU management
Abbreviations: SAH, aneurysmal subarachnoid hemorrhage; ICU, intensive care unit; IRB, institutional review board; ROC, receiver operating characteristic curve; AUC, area under the curve. ⁎ Corresponding author at: 1750-1 Ikenobe, Miki, Kita, Kagawa 761-0793, Japan. E-mail addresses: [email protected] (T. Hifumi), [email protected] (K. Kawakita), [email protected] (H. Shishido), [email protected] (D. Ogawa), [email protected] (M. Okauchi), [email protected] (A. Shindo), [email protected] (M. Kawanishi), [email protected] (S. Inoue), [email protected] (T. Tamiya), [email protected] (Y. Kuroda). 1 Drs. Okazaki and Hifumi contributed equally to this work.
http://dx.doi.org/10.1016/j.jcrc.2017.06.001 0883-9441/© 2017 Elsevier Inc. All rights reserved.
have been reported to be clinically more reliable and accurate than an absolute lactate value as a marker of global tissue hypoxia in various critical conditions [4-6]. Although studies assessing associations between blood lactate level and medical complications in SAH are limited, a study by Satoh et al. reported that increased blood lactate level on admission was correlated with the early onset of neurogenic pulmonary edema in patients with SAH [7]. In addition, the maximal lactate level within 24 h after admission and lactate levels on admission were recently reported to be prognostic factors of unfavorable neurological outcomes in patients with SAH [8,9]; however, serial measurements of lactate levels during ICU stay have not been evaluated in patients with SAH. The purpose of this study was to assess the behavior of serial blood lactate measurements during ICU stay and to identify the prognostic factors of unfavorable neurological outcomes in patients with SAH in the acute phase.
230
T. Okazaki et al. / Journal of Critical Care 41 (2017) 229–233
2. Materials and methods 2.1. Study design and setting Kagawa University Hospital is a 613-bed academic teaching institution with an 8-bed neurointensivist-managed ICU [10]. The protocol of this single-center, retrospective, case-control study, which was performed by retrospective review of medical records, was approved by the institutional review board (IRB) of Kagawa University Hospital and was conducted in accordance with the ethical standards established in the 1964 Declaration of Helsinki and its later amendments. The IRB waived the requirement for patient consent due to the retrospective nature of the study.
Categorical comparisons were conducted using the Fisher's exact test. Univariate and multivariate regression analyses (with step-wise variable selection) were conducted to identify independent factors predictive of unfavorable neurologic outcomes in the study population. Covariates with age, sex (male), Hunt & Kosnik grade, Fisher group, treatment modality (coil or clip), and lactate levels at each time were included in the multivariate analysis. Receiver operating characteristic (ROC) curve and the area under the curve (AUC) were estimated based on the Mann–Whitney U test. For each continuous variable, the cut-off value with the best combination of sensitivity and specificity was identified. Statistical analyses were performed using JMP® statistical software (version 11; SAS Institute, Cary, NC, USA). A twosided p value of b 0.05 was considered statistically significant for all analyses.
2.2. Study participants and inclusion criteria All patients ≥18 years old who were consecutively hospitalized for a confirmed diagnosis of SAH and had at least five arterial lactate measurements between January 1, 2009 and June 30, 2016 were retrospectively enrolled. Patients who received only comfort care within 24 h of admission were excluded. Patients with unknown onset of SAH were also excluded. 2.3. General management of SAH in the ICU Management was conducted in line with the Guidelines for the Management of Aneurysmal Subarachnoid Hemorrhage of the American Heart Association/American Stroke Association [11]. In addition to general intensive care, all patients were monitored for clinical deterioration or cerebral infarction arising due to delayed cerebral ischemia. Fluid management was targeted at maintaining euvolemia. Prophylactic hemodynamic support, hemodilution, and administration of mannitol or hypertonic saline solution were not performed [10,12]. 2.4. Measurement of lactate levels Arterial blood gas analyses, including lactate levels, were routinely performed on admission and every 6 h in the ICU. Additional measurements were carried out as needed by critical care physicians. Arterial blood gas measurements from blood samples collected in heparinized blood gas syringes were measured at 37 °C using a blood gas analyzer [12]. 2.5. Data sampling The following data were collected: age, sex, Hunt and Kosnik grade, Fisher group, treatment modality (coil or clip), serial arterial lactate, modified Rankin scale (mRS) upon hospital discharge, length of ICU stay, length of hospital stay, and hospital mortality. 2.6. Outcome measures The primary outcome was the association of serial lactate levels measured in the ICU with an unfavorable neurologic outcome, which was assessed by the mRS upon hospital discharge. The neurologic outcome was defined as unfavorable when the mRS score was 3–6 and as favorable when the mRS score was 0–2. The secondary outcome was to determine the most accurate predictor in serial lactate measurements. 2.7. Statistical analysis Demographic factors and baseline characteristics were summarized using descriptive statistics. The groups were compared using the Student's t-test or Mann–Whitney U test, as deemed appropriate.
3. Results 3.1. Baseline characteristics of the study population and comparison of baseline characteristics between unfavorable and favorable outcomes The study population included 145 patients (mean age, 62.5 years; 44 men) (Table 1). Neurological outcomes were unfavorable in 46.2% of these patients. Patients with unfavorable neurological outcomes were significantly older than those with favorable outcomes [mean ± standard deviation, 69.5 ± 15.0 vs. 56.5 ± 15.0 years, respectively, p b 0.01]. ICU stay and hospital stay were significantly longer in the unfavorable neurological outcome group than the favorable outcome group [median, interquartile range; 17 (13 − 21) vs. 14 (12–16) days, p b 0.01; 46 (26–75) vs. 24 (21 − 31) days, p b 0.01, respectively].
Table 1 Characteristics of the study population. Variables
All patients (n = 145)
Unfavorable neurological outcomes group (n = 67)
Favorable neurological outcomes group (n = 78)
p value
Age (years) Gender (male) H&K grades I II III IV V Fisher grades 1 2 3 4 Treatment modality Clip Coil Modified Rankin Scale 0 1 2 3 4 5 6 ICU stay Hospital stay Hospital mortality
62.5 ± 16.3 44 (30.3)
69.5 ± 15.0 19 (28.4)
56.5 ± 15.0 25 (32.1)
b0.01 0.63 b0.01
11 (7.6) 55 (37.9) 37 (25.5) 28 (19.3) 14 (9.7)
2 (3.0) 12 (17.9) 21 (31.3) 22 (32.8) 10 (14.9)
9 (11.5) 43 (55.1) 16 (20.5) 6 (7.7) 4 (5.1)
1 (0.7) 34 (23.4) 99 (68.3) 11 (7.6)
0 (0.0) 8 (11.9) 54 (80.6) 5 (7.5)
1 (1.3) 26 (33.3) 45 (57.7) 6 (7.7)
36 (24.8) 109 (75.2)
14 (20.9) 53 (79.1)
22 (28.2) 56 (71.8)
29 (20.0) 22 (15.2) 27 (18.6) 22 (15.2) 24 (16.6) 12 (8.3) 9 (6.2) 15 [12–19] 28 [22–53] 9 (6.2)
0 (0.0) 0 (0.0) 0 (0.0) 22 (32.8) 14 (35.8) 12 (17.9) 9 (13.4) 17 [13−21] 46 [26–75] 9 (13.4)
29 (37.2) 22 (28.2) 27 (34.6) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 14 [12–16] 24 [21−31] 0 (0.0)
b0.01
0.31
b0.01
b0.01 b0.01 b0.01
Data are expressed as mean ± standard deviation, number (percentage), or median [interquartile range]. H&K, Hunt and Kosnik.
T. Okazaki et al. / Journal of Critical Care 41 (2017) 229–233
231
Fig. 1. Serial measurements of lactate in all study patients. The boxes are the 25th to 75th percentiles and the whiskers are the 5th to 95th percentiles.
Serial lactate measurements in all study patients are shown in Fig. 1. Overall, initially increased lactate levels decreased to normal ranges during the first 24 h. 3.2. Alteration of arterial lactate levels between the unfavorable and favorable neurological outcome groups during ICU stay Serial lactate measurements in the unfavorable and favorable neurological outcome groups are shown in Fig. 2. The serum lactate level at each time point was greater in the unfavorable neurological outcome group than the favorable neurological outcome group and initial lactate levels increased to the maximum levels during the first 24 h and then decreased to within normal ranges. Then, the lactate levels slightly increased again to within the normal range for the next 24 h, especially in patients with unfavorable neurological outcome.
3.3. Independent predictors of unfavorable neurologic outcome The results of multivariate logistic regression models are shown in Table 2. Lactate levels measured at 24 h [odds ratio (OR), 2.42; 95% confidence interval (CI), 1.38–4.75; p b 0.01], 36 h (OR, 3.71; 95% CI, 1.80– 8.70; p b 0.01), and 48 h after admission (OR, 3.39; 95% CI, 1.26– 10.81; p = 0.01) were strong predictors of unfavorable neurological outcomes (Supplemental Table 1 demonstrates the result of multivariate logistic regression models).
3.4. ROC curve analysis ROC curves of unfavorable neurological outcomes based on serial lactate levels at each time point were constructed. The respective
Fig. 2. Serial measurements of lactate in the unfavorable and favorable neurological outcome groups. UO, unfavorable neurological outcomes; FO, favorable neurological outcomes. Patients in the unfavorable outcome group are indicated in gray and those in the favorable outcome group are indicated in white. The boxes are the 25th to 75th percentiles and the whiskers are the 5th to 95th percentiles.
232
T. Okazaki et al. / Journal of Critical Care 41 (2017) 229–233
Table 2 Multivariate logistic regression models. Model
Variables
Adjusted odds ratio
95% confidence interval
p value
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
Lactate on admission Lactate at 12 h after admission Lactate at 24 h after admission Lactate at 36 h after admission Lactate at 48 h after admission Lactate at 60 h after admission Lactate at 72 h after admission
1.38 1.33 2.45 3.71 3.41 4.21 1.07
1.02–1.93 0.91–2.00 1.39–4.84 1.80–8.69 1.26–10.93 1.00–21.16 0.34–3.59
b0.05 0.14 b0.01 b0.01 0.01 0.05 0.91
Each model was adjusted according to age, sex (male), Hunt & Kosnik grade, Fisher grade and treatment modality (coil or clip). Age and Hunt & Kosnik grade were significant factors in all models.
AUCs, cut-off levels, sensitivities, and specificities for prediction of unfavorable neurologic outcomes are shown in Table 3. The lactate level measured at 48 h after admission was the most accurate predictor with greater specificity than at other time points (AUC, 0.716; cut-off level, 1.1 mmol/L; sensitivity, 40%; specificity, 92.1%). 3.5. Details of 6 cases with favorable neurological outcome in spite of lactate level N 1.1 mmol/L at 48 h after admission The characteristics of these patients are shown in Supplemental Table 2. In four patients (nos. 1, 2, 4, and 5), extubation was completed by 48 h after admission and these patients complained of severe headache, which was not resolved with analgesics at 48 h. Two patients (nos. 3 and 6) developed serious complications. Patient no. 3 developed both stress-induced cardiomyopathy and neurogenic pulmonary edema, and patient no. 6 was in shock status due to rupture of a pancreaticoduodenal aneurysm. 3.6. Comparison of baseline characteristics between patients with lactate levels at 48 h N 1.1 mmol/L and patients with lactate levels at 48 h ≤ 1.1 mmol/L in the favorable and unfavorable outcome groups The portion of male patients with lactate levels at 48 h ≤ 1.1 mmol/L was significantly higher than that of patients with lactate levels at 48 h N 1.1 mmol/L in the unfavorable outcome group (41% vs. 11.5%, p = 0.01). The same tendency was observed in the favorable outcome group (34.3% vs. 16.7%, p = 0.38) (Supplemental Table 3 demonstrates the result of comparison of baseline characteristics between patients with lactate levels at 48 h N 1.1 mmol/L and patients with lactate levels at 48 h ≤ 1.1 mmol/L in the favorable and unfavorable outcome groups). 3.7. The association between timing of treatment (coil or clip) and alteration of lactate levels during the first 48 h Initial treatment (coil or clip) completed within 30 h after admission in both 95% (138/145) of all study patients and 94% (63/67) of patients with unfavorable neurological outcome. (Supplemental Table 4 demonstrates the result of comparison of timing of initial treatment between the unfavorable neurological outcome group and favorable neurological outcome group in the coil and clip groups).
4. Discussion The main findings of the current study were as follows: (1) in the unfavorable outcome group, initially increased lactate levels decreased to within the normal range for the first 24 h. Then, lactate levels slightly increased again to within the normal range for the next 24 h. (2) The lactate level at 48 h after admission was the most accurate predictor of unfavorable neurological outcomes in the acute phase (from on admission to 72 h after admission) with a specificity N90% in patients with SAH. Serum lactate levels during ICU management of SAH patients can be influenced by multifactorial factors. Not only volume status, but also aerobic glycolysis caused by excessive catecholamine levels and impaired lactate clearance reflecting hepatic dysfunction due to sedatives and antiepileptic drugs [13,14], may increase lactate levels. However, when limited to the first 48 h, an excess in catecholamine is considered the main reason for the increase of lactate levels because hypovolemic shock and hepatic dysfunction were rarely seen in our study populations (data not shown). Ogura et al. examined catecholamine levels during the first 24 h after SAH and observed high catecholamine levels in samples obtained within the first 6 h, but levels in samples obtained around 24 h were mostly normal [15]. Benedict et al. reported that catecholamine levels after neurosurgery for SAH were altered and initial high catecholamine levels were normalized at 12 h (estimated to be approximately 24 h after admission), but were increased again around 24 h after neurosurgery (estimated to be approximately 36 h after admission) [16]. Our current data of alteration of lactate levels in the unfavorable outcome group were paralleled with the above-mentioned alteration of catecholamine levels (Fig. 2). That is, initially increased lactate levels reached maximum levels during the first 24 h and then decreased to within the normal range. Interestingly, lactate levels slightly increased again to within the normal range for the next 24 h. Thus, a distinct difference was observed between the favorable and unfavorable neurological outcome groups at 48 h after admission. Therefore, we believe that the lactate level at 48 h is the most accurate predictor of neurological outcome, as compared to two time points (on admission and maximum level during the first 24 h) as previously reported. van Donkelaar et al. reported an association between the maximal lactate level within 24 h after admission and unfavorable neurological outcomes in patients with SAH [8]. Aisiku et al. reported that elevated
Table 3 Receiver operating characteristic curve analysis of serial measurements of arterial blood lactate levels. Variables
AUC
SE
Cut-off level
Sensitivity
Specificity
Lactate on admission Lactate at 12 h after admission Lactate at 24 h after admission Lactate at 36 h after admission Lactate at 48 h after admission Lactate at 60 h after admission Lactate at 72 h after admission
0.628 0.623 0.676 0.712 0.716 0.695 0.629
0.127 0.160 0.311 0.362 0.540 0.613 0.463
2.9 mmol/L 1.4 mmol/L 1.0 mmol/L 1.0 mmol/L 1.1 mmol/L 0.8 mmol/L 0.9 mmol/L
41.3 55.2 61.2 59.7 40.0 60.9 48.4
82.1 68.0 70.5 77.9 92.1 68.0 72.6
AUC, area under the curve; SE, standard error.
T. Okazaki et al. / Journal of Critical Care 41 (2017) 229–233
serum lactate levels on admission may be predictive of mortality [9]. In the current study, re-evaluation of serial lactate measurements showed that lactate levels at 48 h after admission provided the highest AUC, in accordance with previously reported predictors [8,9]. This may be considered due to the nature of the prognostic factor because, in general, later points can predict more accurate prognosis. However, novel specificity for predicting unfavorable neurological outcome and analysis based on serial measurements of lactate during the first 72 h could be a strength. According to our current data, arterial lactate level measured at 48 h after admission N 1.1 mmol/L is a predictor of unfavorable neurological outcome with a specificity of 92.1%. Lactate measurements are quick, cheap, and more commonly used than other previously reported prognostic markers [17], such as S100β protein [18,19] and neuron-specific enolase [20,21]; therefore, lactate evaluation in itself offers a great advantage. On the other hand, six of 32 cases with lactate level N 1.1 mmol/L at 48 h achieved favorable neurological outcomes with intensive care. Detailed examinations of those cases revealed that patients with complications, such as stress-induced cardiomyopathy and hypovolemia, and severe headache, indicate that a higher catecholamine level may beyond be the cut-off value; therefore, caution is needed when applying these results clinically. According to the comparison of baseline characteristics between patients with lactate levels at 48 h N 1.1 mmol/L vs. ≤ 1.1 mmol/L in the favorable and unfavorable outcome groups, lactate clearance was more dominant in males than in females in both groups. Although the mechanism remains unknown, this tendency should be also kept in mind. We also examined the association between timing of initial treatment and alteration of lactate levels during the first 48 h, especially slightly re-increased lactate levels around 36–42 h in patients with unfavorable neurological outcome. Because 94% of patients with unfavorable neurological outcome completed initial treatment within 30 h, we do not believe that the timing of treatment affected the alteration of lactate levels. This study had several limitations. First, it was a retrospective cohort study conducted at a single center, which introduces potential selection bias. Moreover, uncontrolled confounding factors may exist. Second, it was not possible to follow-up neurological outcomes of patients after discharge. Third, we could not definitively exclude the possibility that changes in the patient population had an impact on neurological outcomes; for example, comorbidities were not examined in the current study. Fourth, recognition of elevation in lactate levels led to the adjustment of fluids and workup to reveal the etiology by critical care physicians; however, that approach was not protocol-based. Finally, the sample size in this study was relatively small. 5. Conclusions Initially increased lactate levels reached maximum levels during the first 24 h and then decreased to within the normal range. Then, lactate levels slightly increased again to within the normal range for the next 24 h in the unfavorable outcome group. The lactate level at 48 h after admission was the most accurate predictor of unfavorable neurological outcomes with a specificity N90% in SAH patients. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jcrc.2017.06.001. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
233
Conflict of interest The authors declare that they have no conflict of interest.
Acknowledgments We are grateful to all physicians and nurses at the study site for their crucial contribution to the successful completion of this study. References [1] Samuels O, Webb A, Culler S, Martin K, Barrow D. Impact of a dedicated neurocritical care team in treating patients with aneurysmal subarachnoid hemorrhage. Neurocrit Care 2011;14:334–40. [2] Josephson SA, Douglas VC, Lawton MT, English JD, Smith WS, Ko NU. Improvement in intensive care unit outcomes in patients with subarachnoid hemorrhage after initiation of neurointensivist co-management. J Neurosurg 2010;112:626–30. [3] Knopf L, Staff I, Gomes J, McCullough L. Impact of a neurointensivist on outcomes in critically ill stroke patients. Neurocrit Care 2012;16:63–71. [4] Bonizzoli M, Lazzeri C, Cianchi G, Boddi M, Cozzolino M, Di Valvasone S, et al. Serial lactate measurements as a prognostic tool in venovenous extracorporeal membrane oxygenation support. Ann Thorac Surg 2016. [5] Bakker J, Gris P, Coffernils M, Kahn RJ, Vincent JL. Serial blood lactate levels can predict the development of multiple organ failure following septic shock. Am J Surg 1996;171:221–6. [6] Vincent JL, Dufaye P, Berre J, Leeman M, Degaute JP, Kahn RJ. Serial lactate determinations during circulatory shock. Crit Care Med 1983;11:449–51. [7] Satoh E, Tagami T, Watanabe A, Matsumoto G, Suzuki G, Onda H, et al. Association between serum lactate levels and early neurogenic pulmonary edema after nontraumatic subarachnoid hemorrhage. J Nippon Med Sch 2014;81:305–12. [8] van Donkelaar CE, Dijkland SA, van den Bergh WM, Bakker J, Dippel DW, Nijsten MW, et al. Early circulating lactate and glucose levels after aneurysmal subarachnoid hemorrhage correlate with poor outcome and delayed cerebral ischemia: a twocenter cohort study. Crit Care Med 2016. [9] Aisiku IP, Chen PR, Truong H, Monsivais DR, Edlow J. Admission serum lactate predicts mortality in aneurysmal subarachnoid hemorrhage. Am J Emerg Med 2016; 34:708–12. [10] Egawa S, Hifumi T, Kawakita K, Okauchi M, Shindo A, Kawanishi M, et al. Impact of neurointensivist-managed intensive care unit implementation on patient outcomes after aneurysmal subarachnoid hemorrhage. J Crit Care 2016;32:52–5. [11] Connolly Jr ES, Rabinstein AA, Carhuapoma JR, Derdeyn CP, Dion J, Higashida RT, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/ American Stroke Association. Stroke 2012;43:1711–37. [12] Okazaki T, Hifumi T, Kawakita K, Shishido H, Ogawa D, Okauchi M, et al. Blood glucose variability: a strong independent predictor of neurological outcomes in aneurysmal subarachnoid hemorrhage. J Intensive Care Med 2016. [13] Bakker J, Nijsten MW, Jansen TC. Clinical use of lactate monitoring in critically ill patients. Ann Intensive Care 2013;3:12. [14] Fuller BM, Dellinger RP. Lactate as a hemodynamic marker in the critically ill. Curr Opin Crit Care 2012;18:267–72. [15] Ogura T, Satoh A, Ooigawa H, Sugiyama T, Takeda R, Fushihara G, et al. Characteristics and prognostic value of acute catecholamine surge in patients with aneurysmal subarachnoid hemorrhage. Neurol Res 2012;34:484–90. [16] Benedict CR, Loach AB. Sympathetic nervous system activity in patients with subarachnoid hemorrhage. Stroke 1978;9:237–44. [17] Hong CM, Tosun C, Kurland DB, Gerzanich V, Schreibman D, Simard JM. Biomarkers as outcome predictors in subarachnoid hemorrhage—a systematic review. Biomarkers 2014;19:95–108. [18] Weiss N, Sanchez-Pena P, Roche S, Beaudeux JL, Colonne C, Coriat P, et al. Prognosis value of plasma S100B protein levels after subarachnoid aneurysmal hemorrhage. Anesthesiology 2006;104:658–66. [19] Sanchez-Pena P, Pereira AR, Sourour NA, Biondi A, Lejean L, Colonne C, et al. S100B as an additional prognostic marker in subarachnoid aneurysmal hemorrhage. Crit Care Med 2008;36:2267–73. [20] Moritz S, Warnat J, Bele S, Graf BM, Woertgen C. The prognostic value of NSE and S100B from serum and cerebrospinal fluid in patients with spontaneous subarachnoid hemorrhage. J Neurosurg Anesthesiol 2010;22:21–31. [21] Mabe H, Suzuki S, Mase M, Umemura A, Nagai H. Serum neuron-specific enolase levels after subarachnoid hemorrhage. Surg Neurol 1991;36:170–4.