Serum Acylcarnitine and Long-Term Functional Prognosis after Traumatic Brain Injury with Intracranial Injury: A Multi-Center Prospective Study

Abstract

Serum biomarkers have potential to help predict prognosis of traumatic brain injury (TBI). The objective of this study was to evaluate the association between serum acylcarnitine levels and functional outcomes at 1 month/6 months after injury for TBI patients with intracranial hemorrhage or diffuse axonal injury. This study is a multi-center prospective cohort study in which adult TBI patients with intracranial injury visiting the emergency departments (EDs) from December 2018 to June 2020 were enrolled. Serum acylcarnitine levels at the time of ED arrival were categorized into four groups: low (1.2-5.5 μmol/L), low-normal (5.6-10.0 μmol/L), high-normal (10.1–14.5 μmol/L), and high (1.4.6–56.6 μmol/L). The study outcome was set as poor functional recovery at 1 month/6 months after injury (Glasgow Outcome Scale score, 1-3). Multi-level logistic regression analyses were conducted to estimate association between serum acylcarnitine and functional outcomes. Among total of 549 patients, poor functional recovery at 1 month and 6 months after injury were observed in 29.1% (160/549) and 29.1% (158/543, follow-up loss n = 6). The odds for 1-month poor functional outcome increased in the high-normal and the high groups [adjusted odds ratios, AORs (95% confidence intervals, CIs): 1.56 (1.09-2.23) and 2.47 (1.63-3.75)], compared with the low-normal group) and also as a continuous variable [1.05 (1.03-1.07) for each 1 μmol/L]. Regarding 6-month mortality, the high group had significantly higher odds when compared with the low-normal group [AOR (95% CI): 2.16 (1.37-3.40)]. Higher serum acylcarnitine levels are associated with poor functional outcomes at 1 month/6 months after injury for TBI patients with intracranial injury.

Introduction

Traumatic brain injury (TBI) is a serious health problem and is a leading cause of death and disability. Approximately 50 million new TBI cases occur annually worldwide, and the mortality rate has been estimated to 30-40%.^1,2 Although the widespread use of brain imaging has helped in diagnosing TBI and predicting neurological outcomes of those patients,³ prediction of prognosis after TBI based solely on radiological findings remains challenging.^4,5 Previous studies have suggested that serum biomarkers such as S100B, glial fibrillary acidic protein (GFAP), and neuron-specific enolase (NSE) have potential to help predict prognosis of patients with TBI.^6
-8

The role of serum acylcarnitine as a prognostic biomarker has previously been reported for out-of-hospital cardiac arrest, sepsis, hemodialysis, and angina pectoris, where elevated acylcarnitine levels were associated with poorer patient prognosis.^9

–12 L-carnitine is a naturally occurring endogenous compound that transports fatty acids into the mitochondria for beta-oxidation.¹³ The process results in esterification of L-carnitine to form acylcarnitine derivatives, where L-carnitine is thereafter present in the human body as both free carnitine and acylcarnitine.¹⁴ In TBI, an increase in serum acylcarnitine could be the result of mitochondrial dysfunction and disruption of cellular metabolism after injury.^15,16 Because of the potential neuroprotective effect of the L-carnitine, previous studies evaluated the effect of administration of exogenous L-carnitine on human and animal cases with TBI.^17,18 However, few studies have investigated the association between serum carnitine levels, including acylcarnitine levels, and clinical outcomes after severe TBI.

We hypothesized that increased serum acylcarnitine levels would be associated with short- and long-term poor functional and survival outcomes in TBI patients with intracranial hemorrhage and diffuse axonal injury. The objective of this study was to evaluate the association between serum acylcarnitine levels and functional and survival outcomes at 1 month/6 months after injury for TBI patients with intracranial injury visiting the emergency department (ED).

Methods

Ethics statement

The study was approved by the Institutional Review Board of all participating hospitals (IRB no.: SNUH-1806-078-951; CNUH-2018-297; KNUH-2018-10-014-007; CBNUH-2018-09-018; BMC-30-2018-85). Informed consent was obtained from the patient or a family member/legal representative of an unconscious patient. Patient information was anonymized prior to analysis.

Study design and setting

We evaluated data from the Pan-Asia Trauma Outcomes Study for Traumatic Brain Injury (PATOS-TBI) study, an ongoing multi-center prospective cohort study that started in 2018 at five participating university hospitals throughout Korea (ClinicalTrials.gov, ID: NCT04718935).¹⁹ The PATOS-TBI study aims to identify nutritional and metabolic biomarkers that are related to prognosis of TBI and to develop a prediction model of good prognosis with long-term follow-up funded by the National Research Foundation of Korea.

The PATOS-TBI study prospectively enrolls adult TBI patients who were transported by emergency medical services (EMS) to participating hospitals' ED within 72 h after injury and who had intracranial injury defined as intracranial hemorrhage or diffuse axial injury. Based on radiological examination in the ED including computer tomography (CT) and magnetic resonance imaging (MRI), TBI patients who were diagnosed with intracranial injury were included after informed consent from the patients or patients' next of kin in cases of unconsciousness. Patients with penetrating brain injury, those who have had pre-diagnosed neurological diseases, psychological diseases, or terminal cancer, those who were pregnant, and patients transferred to participating hospitals' ED after surgery at other hospitals were excluded from the study.

The PATOS-TBI study registry included patients' demographics; injury characteristics; clinical findings including vital signs, neurological examination results, laboratory test results, brain imaging findings, diagnoses, and in-hospital treatment; and survival/functional outcomes at time of hospital discharge and follow-up. Upon confirmation of intracranial injury and consent to enrollment to the study, blood samples of biomarkers were obtained for analysis in the ED. Patient long-term outcome variables were investigated by telephone interview, and survival/functional outcomes at 1 and 6 months after injury were collected. For quality assurance of the PATOS-TBI registry, a trained data entry researcher was designated for each participating hospital, and quality control of entered data was performed by the PATOS-TBI data quality control team on a monthly basis.¹⁹

Study population

The study enrolled adult TBI patients 18 years or older between December 2018 to June 2020. Patients with TBI diagnosed with intracranial injury confirmed by brain CT or MRI were included. Patients with unknown information on the serum acylcarnitine level or the Glasgow Outcome Scale (GOS) score at 1-month follow-up were excluded.

Outcome measures

The main study outcomes were functional recovery at 1 month and 6 months after injury. Poor functional outcome was defined as GOS score of 1 (death), 2 (persistent vegetative state), or 3 (severe disability).²⁰ The secondary outcomes were functional recovery at hospital discharge, and in-hospital, and 1-month and 6-month mortality.

Analysis of serum biomarkers

Upon confirmation of intracranial injury on brain imaging and patient consent to study enrollment in the ED, 24 mL of blood was drawn via venipuncture. Within 60 min after having been drawn, the samples were centrifuged for 10 min at 3000 rpm at room temperature for separation of serum. The separated serum was frozen at -20°C and analyzed within 7 days for serum biomarkers.

Serum acylcarnitine level was calculated by subtracting free carnitine level from total carnitine level. Serum total carnitine and free carnitine levels were measured by liquid chromatography with electrospray ionization tandem mass spectrometry.²¹ To 150 μL of the aliquoted serum, 30 μL of 400 μmol/L solution of internal standard D3-carnitine was added. Followed by vortex mixing, 30μL of the mixture was applied to cotton fiber filter paper to be used for free carnitine quantification. The remaining mixture was hydrolyzed with 10 μL of 1 mol/L KOH and incubated at 65°C for 15 min, followed by neutralization by 10 μL of 1 mol/L HCl. The mixture was then applied to cotton fiber filter paper for quantification of total carnitine. The specimens were incubated at 40°C for 30 min and extracted with 200 μL of 80% methanol. Aliquots of the samples (5 μL) were injected into the electrospray ion source of the tandem mass spectrometer for data acquisition. Serum concentrations of carnitine was calculated using the ratio of signals of carnitine to its internal standard (carnitine, m/z 162 → m/z 85; internal standard, m/z 165 → m/z 85).

Exposure and variables

The main exposure was serum acylcarnitine level, which was categorized into four groups according to reference range of acylcarnitine (5.6-14.4 μmol/L) derived from healthy adults reported in a previous study: low (1.2-5.5 μmol/L), low-normal (5.6-10.0 μmol/L), high-normal (10.1-14.5 μmol/L), and high (14.6-56.6 μmol/L).²²

We included data regarding patients' demographics (age, sex, body mass index [BMI]), comorbidities (diabetes mellitus, renal disease, hemodialysis, liver disease, heart failure, cancer, and coagulation abnormality), injury characteristics (mechanism of injury and injury severity score), clinical findings (initial vital signs and Glasgow Coma Scale [GCS] scores at ED, type of intracranial injury found on brain imaging, need for surgery, and disposition after ED treatment), and patients' outcomes at time of hospital discharge and follow-up.

Statistical analysis

Descriptive analysis was conducted to compare the characteristics of study population. Categorical variables were reported as number (percentage) and were compared using the chi-squared test or Fisher's exact test. Continuous variables were reported as median (interquartile range [IQR]) and compared using the Wilcoxon rank sum test.

Multi-level logistic regression analysis was used to investigate whether serum acylcarnitine level was related to the study outcomes, after adjusting for hospital clustering. The adjusted odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated for study outcomes after adjusting for age, sex, obesity (BMI ≥25), diabetes mellitus, renal disease or hemodialysis, other comorbidities, and injury severity score.²³

Sensitivity analyses were conducted for study population groups based on time period between the injury and the blood sample collection during which time-dependent changes in acylcarnitine levels could have occurred, to assess whether the associations between serum acylcarnitine level and study outcomes were maintained in patients for whom blood samples were collected within 24, 12, 6, and 3 h after TBI.²⁴

All tests were two-tailed and p values of <0.05 were considered statistically significant. All statistical analyses were performed using the SAS software (version 9.4; SAS Institute Inc., Cary, NC).

Results

Among 606 patients with intracranial injury in the PATOS-TBI study, 549 patients fulfilled the inclusion criteria, excluding cases with unknown information on the serum acylcarnitine level (n = 1) and the GOS score at 1 month after injury (n = 56). The characteristics of the excluded population with unknown GOS scores at 1-month can be found in Supplementary Table S1.

Table 1 presents the patients' characteristics and outcomes according to level of serum acylcarnitine. The median (IQR) of serum acylcarnitine was 12.5 (8.5-17.1) μmol/L and time from injury to blood sample collection was 3.0 (1.5-5.7) h. Poor functional recovery at 1 month after injury was observed in 29.1% (160/549), and 1-month mortality was 18.8% (103/549). Among 543 patients who were followed up to 6 months after injury, 29.1% (158/543) had poor functional recovery and 20.6% (112/543) died. The proportions of poor functional recovery at 6 months after injury were 28.9% (11/38) for the low group, 21.2% (31/146) for the low-normal group, 25.9% (42/162) for the high-normal group, and 37.4% (76/203) for the high group (p = 0.096).

Table 1.

Characteristics of Study Population by Serum Acylcarnitine Level

	Total	Serum acylcarnitine level (μmol/L)
		Low	Low-normal	High-normal	High	p value
		(1.2-5.5)	(5.6-10.0)	(10.1-14.5)	(14.6-56.6)
	N (%)	N (%)	N (%)	N (%)	N (%)
Total	549	38	146	162	203
Acylcarnitine, μmol/L, median (IQR)	12.5 (8.5-17.1)	4.7 (3.6-5.0)	7.8 (7.0-8.8)	12.3 (11.3-13.4)	18.8 (16.7-23.3)	< 0.001
Total carnitine, μmol/L, median (IQR)	49.6 (39.7-61.1)	38.0 (26.6-44.3)	42.4 (34.8-49.7)	48.0 (39.7-58.4)	60.0 (50.7-74.0)	< 0.001
Free carnitine, μmol/L, median (IQR)	36.8 (28.7-45.9)	34.4 (21.2-40)	34.8 (26.9-42.1)	35.9 (27.7-45.8)	40.0 (32.0-50.3)	< 0.001
Time from injury to blood sample collection, h, median (IQR)	3.0 (1.5-5.7)	3.6 (2.3-6.2)	2.8 (1.4-5.7)	3.5 (1.8-6.9)	2.8 (1.5-4.7)	0.107
Demographics
Age, year, mean (SD)	64.4 (16.9)	71.9 (13.5)	67.7 (15.8)	63.5 (16.6)	61.2 (17.7)	< 0.001
Age, ≥65 years old	305 (55.6)	36 (94.7)	103 (70.5)	82 (50.6)	84 (41.4)	< 0.001
Male sex	376 (68.5)	20 (52.6)	95 (65.1)	103 (63.6)	158 (77.8)	0.002
Body mass index, ≥25 kg/m²	114 (20.8)	7 (18.4)	21 (14.4)	37 (22.8)	49 (24.1)	0.144
Education, >12 years	100 (18.2)	2 (5.3)	23 (15.8)	32 (19.8)	43 (21.2)	0.107
Comorbidities
Diabetes mellitus	141 (25.7)	14 (36.8)	41 (28.1)	41 (25.3)	45 (22.2)	0.236
Renal disease	34 (6.2)	2 (5.3)	8 (5.5)	11 (6.8)	13 (6.4)	0.960
Hemodialysis	20 (3.6)	2 (5.3)	4 (2.7)	8 (4.9)	6 (3.0)	0.643
Liver disease	24 (4.4)	0 (0.0)	7 (4.8)	9 (5.6)	8 (3.9)	0.488
Heart failure	11 (2.0)	0 (0.0)	4 (2.7)	3 (1.9)	4 (2.0)	0.753
Cancer	23 (4.2)	2 (5.3)	6 (4.1)	5 (3.1)	10 (4.9)	0.831
Coagulation abnormality	13 (2.4)	2 (5.3)	6 (4.1)	2 (1.2)	3 (1.5)	0.180
Mechanism of injury						0.643
Road traffic injury	223 (40.6)	12 (31.6)	65 (44.5)	59 (36.4)	87 (42.9)
Fall/slip down	238 (43.4)	18 (47.4)	62 (42.5)	74 (45.7)	84 (41.4)
Others	88 (16.0)	8 (21.1)	19 (13.0)	29 (17.9)	32 (15.8)
Injury severity score
Score, median (IQR)	22 (13-29)	21.5 (16-29)	17 (12-25)	22 (13-29)	22 (14-30)	0.274
Score, ≥16	389 (70.9)	29 (76.3)	100 (68.5)	112 (69.1)	148 (72.9)	0.662
Initial ED vital signs, median (IQR)
Systolic blood pressure, mmHg	133 (115-152)	140 (130-160)	140 (120-160)	130 (110-150)	131.5 (110-150)	0.024
Diastolic blood pressure, mmHg	80 (70-90)	80 (70-90)	80 (70-93)	80 (70-90)	80 (69-90)	0.100
Heart rate, per min	84 (72.5-96)	80 (73-86)	80.5 (71-97)	80 (70.5-92)	88 (76-100)	< 0.001
Respiratory rate, per min	20 (18-20)	20 (18-20)	20 (18-20)	20 (18-20)	20 (20-20)	0.020
Oxygen saturation, %	97 (96-98)	97 (95-98)	97 (96-98)	97 (96-98)	97 (96-98)	0.571
Initial GCS						0.006
3-8	121 (22.0)	6 (15.8)	19 (13.0)	36 (22.2)	60 (29.6)
9-13	71 (12.9)	4 (10.5)	22 (15.1)	19 (11.7)	26 (12.8)
14-15	344 (62.7)	28 (73.7)	104 (71.2)	106 (65.4)	106 (52.2)
Score, mean (SD)	12.1 (4.3)	13.1 (3.6)	13.1 (3.5)	12.2 (4.2)	11.1 (4.7)	< 0.001
Intracranial injury
Subdural hemorrhage	411 (74.9)	30 (78.9)	114 (78.1)	115 (71.0)	152 (74.9)	0.487
Subarachnoid hemorrhage	215 (39.2)	13 (34.2)	50 (34.2)	70 (43.2)	82 (40.4)	0.374
Intracerebral hemorrhage	121 (22.0)	12 (31.6)	31 (21.2)	34 (21.0)	44 (21.7)	0.534
Epidural hemorrhage	80 (14.6)	5 (13.2)	10 (6.8)	33 (20.4)	32 (15.8)	0.009
Diffuse axonal injury	36 (6.6)	6 (15.8)	7 (4.8)	11 (6.8)	12 (5.9)	0.103
Intraventricular hemorrhage	47 (8.6)	4 (10.5)	10 (6.8)	14 (8.6)	19 (9.4)	0.825
Operation
Any operation	160 (29.1)	7 (18.4)	42 (28.8)	49 (30.2)	62 (30.5)	0.491
Neurosurgery	109 (19.9)	6 (15.8)	30 (20.5)	30 (18.5)	43 (21.2)	0.839
ICU admission	330 (60.1)	19 (50.0)	77 (52.7)	107 (66.0)	127 (62.6)	0.050
Survival outcomes
In-hospital mortality	56 (10.2)	2 (5.3)	7 (4.8)	10 (6.2)	37 (18.2)	< 0.001
1-month mortality	103 (18.8)	6 (15.8)	19 (13.0)	24 (14.8)	54 (26.6)	0.004
6-month mortality (n = 543)	112 (20.6)	7 (18.4)	21 (14.7)	26 (16.4)	58 (29.0)	0.003
GOS score at discharge						< 0.001
1	56 (10.2)	2 (5.3)	7 (4.8)	10 (6.2)	37 (18.2)
2	23 (4.2)	2 (5.3)	4 (2.7)	7 (4.3)	10 (4.9)
3	59 (10.7)	5 (13.2)	9 (6.2)	22 (13.6)	23 (11.3)
4	57 (10.4)	3 (7.9)	20 (13.7)	13 (8.0)	21 (10.3)
5	354 (64.5)	26 (68.4)	106 (72.6)	110 (67.9)	112 (55.2)
Poor functional recovery at discharge	138 (25.1)	9 (23.7)	20 (13.7)	39 (24.1)	70 (34.5)	< 0.001
GOS score at 1 month after injury						0.044
1	103 (18.8)	6 (15.8)	19 (13.0)	24 (14.8)	54 (26.6)
2	8 (1.5)	0 (0.0)	1 (0.7)	4 (2.5)	3 (1.5)
3	49 (8.9)	5 (13.2)	11 (7.5)	14 (8.6)	19 (9.4)
4	43 (7.8)	3 (7.9)	11 (7.5)	12 (7.4)	17 (8.4)
5	346 (63.0)	24 (63.2)	104 (71.2)	108 (66.7)	110 (54.2)
Poor functional recovery at 1 month	160 (29.1)	11 (28.9)	31 (21.2)	42 (25.9)	76 (37.4)	0.096
GOS score at 6 months after injury						0.142
1	112 (20.4)	7 (18.4)	21 (14.4)	26 (16.0)	58 (28.6)
2	9 (1.6)	1 (2.6)	3 (2.1)	4 (2.5)	1 (0.5)
3	37 (6.7)	4 (10.5)	8 (5.5)	11 (6.8)	14 (6.9)
4	38 (6.9)	1 (2.6)	9 (6.2)	13 (8.0)	15 (7.4)
5	347 (63.2)	25 (65.8)	103 (70.5)	107 (66.0)	112 (55.2)
Follow-up loss	6 (1.1)	0 (0.0)	2 (1.4)	1 (0.6)	3 (1.5)
Poor functional recovery at 6 months	158 (29.1)	12 (31.6)	32 (22.4)	41 (25.8)	73 (36.5)	0.021

IQR, interquartile range; SD, standard deviation; ED, emergency department; GCS, Glasgow Coma Scale; ICU, intensive care unit; GOS, Glasgow Outcome Scale.

In Figure 1, patients with poor functional recovery at 1 month after injury had significantly higher serum acylcarnitine levels, compared with patients with favorable functional outcomes (15.8 vs. 13.2 μmol/L (p < 0.001). The acylcarnitine cutoff value of 10.0 μmol/L resulted in sensitivity of 75.2% and specificity of 35.7% while cutoff value of 14.5 μmol/ L resulted in sensitivity of 47.4% and specificity of 66.5%.

FIG. 1.

Kernel density plots of serum acylcarnitine according to functional recovery at 1-month after injury. Poor functional outcome defined as Glasgow outcome scale score of 1 (death), 2 (persistent vegetative state), or 3 (severe disability).

Table 2 presents the associations between serum acylcarnitine levels and study outcomes. The adjusted odds for poor functional recovery at 1 month after injury increased in the high-normal and the high groups [adjusted odds ratios, ORs (95% confidence intervals, CIs)] when compared with the low-normal group: 1.56 (1.09-2.23) for the high-normal group and 2.47 (1.63-3.75) for the high group). When acylcarnitine was analyzed as a continuous variable, each 1 μmol/L increase in serum acylcarnitine level was associated with significant increase in odds for poor 1-month functional outcome [adjusted OR (95% CI): 1.05 (1.03-1.07)]. When 543 patients who were followed up to 6 months were analyzed, the high group had statistically significantly higher odds of poor functional recovery at 6-months after injury [adjusted OR (95% CI): 2.16 (1.37-3.40)] and 6-month mortality [adjusted OR (95% CI): 2.86 (1.91-4.28)] when compared with the low-normal group.

Table 2.

Multi-Level Logistic Regression Analysis of Serum Acylcarnitine level Category for Study Outcomes

Outcomes	Outcome, n/N (%)	Model 1	Model 2	Model 3
Outcomes	Outcome, n/N (%)	OR (95% CI)	Adjusted OR (95% CI)	Adjusted OR (95% CI)
Poor functional recovery at 1 month after injury
Low (1.2-5.5)	11/38 (28.9)	1.09 (0.71-1.66)	0.95 (0.59-1.53)	0.90 (0.56-1.45)
Low-normal (5.6-10.0)	31/146 (21.2)	Reference	Reference	Reference
High-normal (10.1-14.5)	42/162 (25.9)	1.36 (0.97-1.91)	1.54 (1.05-2.25)	1.56 (1.09-2.23)
High (14.6-56.6)	76/203 (37.4)	2.09 (1.53-2.85)	2.45 (1.65-3.65)	2.47 (1.63-3.75)
Continuous (per 1 μmol/L increase)	160/549 (29.1)	1.04 (1.03-1.05)	1.05 (1.04-1.06)	1.05 (1.03-1.07)
Poor functional recovery at 6 months after injury
Low (1.2-5.5)	12/38 (31.6)	1.14 (0.59-2.20)	1.03 (0.57-1.87)	0.99 (0.52-1.86)
Low-normal (5.6-10.0)	32/144 (22.2)	Reference	Reference	Reference
High-normal (10.1-14.5)	41/161 (25.5)	1.18 (0.80-1.76)	1.36 (0.82-2.27)	1.37 (0.84-2.25)
High (14.6-56.6)	73/200 (36.5)	1.87 (1.36-2.57)	2.16 (1.39-3.35)	2.16 (1.37-3.40)
Continuous (per 1 μmol/L increase)	158/543 (29.1)	1.04 (1.03-1.06)	1.05 (1.04-1.06)	1.05 (1.04-1.07)
Poor functional recovery at discharge
Low (1.2-5.5)	9/38 (23.7)	1.18 (0.68-2.06)	1.06 (0.63-1.78)	1.00 (0.56-1.76)
Low-normal (5.6-10.0)	20/146 (13.7)	Reference	Reference	Reference
High-normal (10.1-14.5)	39/162 (24.1)	2.00 (1.28-3.13)	2.40 (1.35-4.24)	2.39 (1.41-4.04)
High (14.6-56.6)	70/203 (34.5)	2.96 (1.99-4.39)	3.64 (2.26-5.88)	3.61 (2.25-5.80)
Continuous (per 1 μmol/L increase)	138/549 (25.1)	1.05 (1.04-1.06)	1.06 (1.05-1.07)	1.06 (1.05-1.08)
In-hospital mortality
Low (1.2-5.5)	2/38 (5.3)	0.55 (0.38-0.81)	0.48 (0.29-0.80)	0.44 (0.29-0.69)
Low-normal (5.6-10.0)	7/146 (4.8)	Reference	Reference	Reference
High-normal (10.1-14.5)	10/162 (6.2)	1.12 (0.37-3.41)	1.41 (0.44-4.51)	1.37 (0.43-4.32)
High (14.6-56.6)	37/203 (18.2)	3.36 (2.04-5.53)	4.61 (2.19-9.68)	4.48 (2.21-9.07)
Continuous (per 1 μmol/L increase)	56/549 (10.2)	1.06 (1.03-1.10)	1.08 (1.06-1.10)	1.08 (1.06-1.10)
1-month mortality
Low (1.2-5.5)	6/38 (15.8)	0.87 (0.58-1.29)	0.71 (0.44-1.14)	0.67 (0.43-1.05)
Low-normal (5.6-10.0)	19/146 (13.0)	Reference	Reference	Reference
High-normal (10.1-14.5)	24/162 (14.8)	1.12 (0.69-1.83)	1.40 (0.81-2.40)	1.41 (0.84-2.38)
High (14.6-56.6)	54/203 (26.6)	2.03 (1.54-2.68)	2.71 (1.71-4.30)	2.74 (1.76-4.27)
Continuous (per 1 μmol/L increase)	103/549 (18.8)	1.04 (1.02-1.05)	1.05 (1.04-1.06)	1.05 (1.04-1.06)
6-month mortality
Low (1.2-5.5)	7/38 (18.4)	0.79 (0.54-1.16)	0.65 (0.42-1.02)	0.62 (0.40-0.95)
Low-normal (5.6-10.0)	21/144 (14.6)	Reference	Reference	Reference
High-normal (10.1-14.5)	26/161 (16.1)	1.14 (0.73-1.80)	1.40 (0.83-2.34)	1.41 (0.86-2.33)
High (14.6-56.6)	58/200 (29.0)	2.13 (1.66-2.73)	2.81 (1.83-4.30)	2.86 (1.91-4.28)
Continuous (per 1 μmol/L increase)	112/543 (20.6)	1.05 (1.02-1.07)	1.06 (1.05-1.08)	1.06 (1.05-1.08)

Model 1: unadjusted logistic regression.

Model 2: adjusted for hospital, demographics (age, sex, and obesity) and comorbidities (diabetes mellitus, renal disease or hemodialysis, and other comorbidities).

Model 3: adjusted for variables in the Model 2, and injury severity score.

OR, odds ratio; CI, confidence interval.

Sensitivity analysis

Among total population, 494 (90.0%) patients with intracranial injury had blood sample collected within 24 h after TBI. Compared with the low-normal acylcarnitine group, the high-normal and the high groups had higher odds for poor functional recovery at 1 month [adjusted ORs (95% CIs): 1.68 (1.24-2.28) and 2.58 (1.78-3.73)] as well as 1-month mortality [adjusted ORs (95% CIs): 1.73 (1.06-2.84) and 3.53 (2.41-5.17); Table 3]. These associations showed a similar trend in patients with blood samples collected within 12 (n = 463), 6 (n = 420), and 3 (n = 281) h after injury.

Table 3.

Sensitivity Analysis for Study Population Based on Time Periods between the Injury and the Blood Sample Collection

Outcomes	Outcome, n/N (%)	Model 1	Model 2	Model 3
Outcomes	Outcome, n/N (%)	OR (95% CI)	Adjusted OR (95% CI)	Adjusted OR (95% CI)
Sample collection within 24 h after injury (N = 494)
Poor functional recovery at 1-months after injury
Low (1.2-5.5)	8/33 (24.2)	1.02 (0.59-1.77)	0.88 (0.48-1.60)	0.83 (0.44-1.57)
Low-normal (5.6-10.0)	27/130 (20.8)	Reference	Reference	Reference
High-normal (10.1-14.5)	40/145 (27.6)	1.50 (1.10-2.05)	1.69 (1.22-2.34)	1.68 (1.24-2.28)
High (14.6-56.6)	72/186 (38.7)	2.19 (1.59-3.03)	2.59 (1.87-3.58)	2.58 (1.78-3.73)
Continuous (per 1 μmol/L increase)	147/494 (29.8)	1.04 (1.03-1.05)	1.05 (1.04-1.06)	1.05 (1.04-1.07)
1-month mortality
Low (1.2-5.5)	4/33 (12.1)	1.09 (0.62-1.93)	0.85 (0.45-1.60)	0.82 (0.44-1.54)
Low-normal (5.6-10.0)	15/130 (11.5)	Reference	Reference	Reference
High-normal (10.1-14.5)	23/145 (15.9)	1.37 (0.84-2.23)	1.72 (1.04-2.83)	1.73 (1.06-2.84)
High (14.6-56.6)	53/186 (28.5)	2.52 (1.90-3.35)	3.49 (2.41-5.06)	3.53 (2.41-5.17)
Continuous (per 1 μmol/L increase)	95/494 (19.2)	1.04 (1.02-1.07)	1.06 (1.04-1.07)	1.06 (1.05-1.07)
Sample collection within 12 h after injury (N = 463)
Poor functional recovery at 1-months after injury
Low (1.2-5.5)	8/31 (25.8)	1.07 (0.64-1.77)	0.93 (0.54-1.60)	0.87 (0.47-1.59)
Low-normal (5.6-10.0)	26/124 (21.0)	Reference	Reference	Reference
High-normal (10.1-14.5)	37/137 (27.0)	1.45 (1.09-1.94)	1.67 (1.22-2.31)	1.65 (1.21-2.26)
High (14.6-56.6)	68/171 (39.8)	2.36 (1.69-3.28)	2.79 (1.97-3.93)	2.80 (1.94-4.03)
Continuous (per 1 μmol/L increase)	139/463 (30.0)	1.04 (1.03-1.05)	1.05 (1.04-1.06)	1.05 (1.04-1.07)
1-month mortality
Low (1.2-5.5)	4/31 (12.9)	1.18 (0.72-1.91)	0.93 (0.55-1.59)	0.89 (0.49-1.59)
Low-normal (5.6-10.0)	14/124 (11.3)	Reference	Reference	Reference
High-normal (10.1-14.5)	21/137 (15.3)	1.32 (0.80-2.16)	1.69 (1.03-2.76)	1.70 (1.03-2.82)
High (14.6-56.6)	52/171 (30.4)	2.86 (2.15-3.81)	3.93 (2.69-5.75)	4.03 (2.78-5.85)
Continuous (per 1 μmol/L increase)	91/463 (19.7)	1.04 (1.02-1.07)	1.06 (1.05-1.08)	1.06 (1.05-1.07)
Sample collection within 6 h after injury (N = 420)
Poor functional recovery at 1-months after injury
Low (1.2-5.5)	7/28 (25.0)	0.99 (0.54-1.82)	0.89 (0.47-1.68)	0.85 (0.42-1.73)
Low-normal (5.6-10.0)	24/114 (21.1)	Reference	Reference	Reference
High-normal (10.1-14.5)	31/116 (26.7)	1.49 (1.11-2.00)	1.69 (1.14-2.49)	1.68 (1.18-2.38)
High (14.6-56.6)	67/162 (41.4)	2.55 (1.73-3.78)	2.98 (2.11-4.22)	2.96 (2.01-4.37)
Continuous (per 1 μmol/L increase)	129/420 (30.7)	1.05 (1.04-1.07)	1.06 (1.06-1.07)	1.06 (1.05-1.07)
1-month mortality
Low (1.2-5.5)	3/28 (10.7)	0.97 (0.55-1.72)	0.80 (0.44-1.42)	0.76 (0.40-1.47)
Low-normal (5.6-10.0)	13/114 (11.4)	Reference	Reference	Reference
High-normal (10.1-14.5)	18/116 (15.5)	1.41 (0.89-2.26)	1.81 (1.14-2.89)	1.88 (1.19-2.96)
High (14.6-56.6)	51/162 (31.5)	2.97 (2.09-4.23)	4.16 (2.73-6.35)	4.23 (2.75-6.53)
Continuous (per 1 μmol/L increase)	85/420 (20.2)	1.05 (1.02-1.09)	1.07 (1.05-1.09)	1.07 (1.06-1.09)
Sample collection within 3 h after injury (N = 281)
Poor functional recovery at 1-months after injury
Low (1.2-5.5)	4/14 (28.6)	1.12 (0.41-3.06)	0.96 (0.37-2.48)	0.87 (0.33-2.31)
Low-normal (5.6-10.0)	17/78 (21.8)	Reference	Reference	Reference
High-normal (10.1-14.5)	23/74 (31.1)	1.79 (1.31-2.43)	2.00 (1.39-2.89)	1.89 (1.34-2.68)
High (14.6-56.6)	47/115 (40.9)	2.24 (1.67-3.01)	2.71 (1.90-3.89)	2.56 (1.71-3.85)
Continuous (per 1 μmol/L increase)	91/281 (32.4)	1.04 (1.02-1.05)	1.05 (1.03-1.06)	1.05 (1.03-1.06)
1-month mortality
Low (1.2-5.5)	1/14 (7.1)	0.67 (0.40-1.11)	0.53 (0.34-0.81)	0.49 (0.32-0.75)
Low-normal (5.6-10.0)	8/78 (10.3)	Reference	Reference	Reference
High-normal (10.1-14.5)	14/74 (18.9)	2.14 (1.28-3.56)	2.81 (1.67-4.71)	2.71 (1.59-4.62)
High (14.6-56.6)	35/115 (30.4)	3.06 (2.08-4.50)	4.50 (2.53-7.98)	4.25 (2.35-7.69)
Continuous (per 1 μmol/L increase)	58/281 (20.6)	1.04 (1.00-1.08)	1.06 (1.03-1.09)	1.06 (1.03-1.08)

Model 1: unadjusted logistic regression.

Model 2: adjusted for hospital, demographics (age, sex, and obesity) and comorbidities (diabetes mellitus, renal disease or hemodialysis, and other comorbidities).

Model 3: adjusted for variables in the Model 2, and injury severity score.

OR, odds ratio; CI, confidence interval.

Discussion

Using the multi-center prospective PATOS-TBI registry, this study evaluated the association between serum acylcarnitine levels and short- and long-term functional/survival prognoses of adult TBI patients with intracranial hemorrhage or diffuse axonal injury. There was an association between serum acylcarnitine levels and 1-month poor functional recovery [adjusted OR (95% CI) when compared with the low-normal group: 1.56 (1.09-2.23) for the high-normal group and 2.47 (1.63-3.75) for the high group]. With regard to long-term functional outcome at 6 months, there was an association between serum acylcarnitine levels and poor functional recovery [adjusted OR (95% CI) when compared with the low-normal group: 2.16 (1.37-3.40) for the high group]. A similar trend was observed in patients with blood samples collected within 24 (n = 494), 12 (n = 463), 6 (n = 420) and 3 (n = 281) h after injury. These results suggest serum acylcarnitine as a potential biomarker related to the prognosis of TBI with intracranial injury.

Carnitine is an essential factor in cellular metabolism and is also known to have neuroprotective effects.^25,26 Acylcarnitine derivatives are formed by esterification of L-carnitine during the process of transporting fatty acids into the mitochondria for beta-oxidation.¹³ Previous human and animal studies have reported that severe TBI triggers hypermetabolic state of the brain which increases glucose utilization.^27,28 Enhanced fatty acid oxidation to supply energy for the brain is resulted in a decrease in free carnitine and increase of acylcarnitine.^26,29,30 Among serum acylcarnitines, long-chain acylcarnitines (palmitoylcarnitine and oleoylcarnitine) were overexpressed in brain tissues only after injury in rats with TBI, which suggested acylcarnitine as injury-related biomarkers.^31,32 In another study, a combination of branched-chain amino acid metabolites (2-methylbutyrylcarnitine, propionylcarnitine and 4-methyl-2-oxopentanoate) was utilized for predicting increased intracranial pressure in patients with severe TBI.³³ As shown, the results of this study that higher levels of serum acylcarnitine are associated with poor prognosis is in line with previous studies.

In Korea, more than 100,000 TBI patients visit the ED every year, and about 4000 patients die due to TBI.³⁴ Determining the prognosis of severe TBI patients based solely on brain imaging and demographic factors has limitations,^3
-5 and previous studies have suggested that serum biomarkers could be helpful in predicting prognostication.^1,6
-8 Acylcarnitine profile has been traditionally utilized for screening of inborn errors such as long-chain acyl-CoA dehydrogenase deficiency and propionyl-CoA carboxylase deficiency.²⁵ In this study, serum acylcarnitine could be utilized as a biomarker for prognostication of TBI patients with intracranial hemorrhage or diffuse axonal injury. As has been previously reported for biomarkers S100B, GFAP, and NSE,^6
-8 further studies are needed investigating the efficacy of serum acylcarnitine in combination with other biomarkers in assisting prediction of functional/survival outcomes of patient with TBI.^35,36

This study has several limitations. First, the blood-based protein biomarker concentration would be affected by elapsed time from injury. However, blood samples in this study were collected only at the time of ED arrival and therefore missing data on kinetics of serum acylcarnitine level. The kinetics of acylcarnitine over time after TBI is yet to be studied and further studies considering serial follow up are needed. Second, this study only evaluated total acylcarnitine levels and not specific carnitine components, although serum acylcarnitine is composed of short-, medium-, and long-chain acylcarnitine that have different chemical and metabolic properties.^12,14,25 Third, there was no available data for diet diaries and/or nutritional status, which would be related to serum carnitine levels.³⁷ Fourth, the acylcarnitine levels could have been affected by accompanied trauma other than brain injury such as orthopedic, spinal cord, or myocardial injury,^9,33,38 which was not considered in the analysis and were only adjusted with injury severity score. Fifth, even though the study population was classified according to the reference range of acylcarnitine level in healthy adults, the study lacks serum acylcarnitine levels in age- and sex-matched controls. Lastly, selection bias may arise because the patients with unknown GOS scores at 1 month were excluded. Caution should be taken when interpreting this study results given the significant limitations.

Conclusions

The higher serum acylcarnitine levels are associated with short- and long-term poor functional and survival outcomes for TBI patients with intracranial injury. This study suggests that serum acylcarnitine may be a potential biomarker for predicting the prognosis of patients with TBI with intracranial hemorrhage or diffuse axonal injury when utilized in combination with other biomarkers.

Footnotes

Acknowledgments

Data availability: The data for the study were obtained from the National Research Foundation of Korea (NRF) but restrictions apply to the availability of these data and so are not publicly available.

Ethics statement: The study protocol was reviewed and approved by the Institutional Review Board of all participating hospitals (IRB no.: SNUH-1806-078-951; CNUH-2018-297; KNUH-2018-10-014-007; CBNUH-2018-09-018; BMC-30-2018-85).

Authors' Contributions

Drs. Ro and Lee had full access to all of the data in the study and take responsibility for the integrity of the data as well as the accuracy of the data analysis.

Conceptualization: Dr. Lee and Dr. Ro.

Data curation: Dr. Lee, Dr. Jung, Dr. Moon, Dr. Park (GJP), Dr. Yoon, and Dr. Park (JHP).

Formal analysis: Dr. Lee.

Funding acquisition: Dr. Ro.

Investigation: Dr. Ro, Dr. Jung, Dr. Moon, Dr. Park (GJP), and Dr. Yoon.

Methodology: Dr. Lee, and Dr. Ro, and Dr. Shin.

Software: Dr. Lee.

Supervision: Dr. Ro, Dr. Park (JHP), and Dr. Shin.

Validation: Dr. Jung, Dr. Moon, Dr. Park (GJP), and Dr. Yoon.

Visualization: Dr. Lee.

Writing—original draft: Dr. Lee.

Writing—review and editing: Dr. Ro and Dr. Shin.

Approval of final manuscript: all authors.

Funding Information

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (Grant no.: NRF-2018R1C1B6007625, NRF-2021R1A2C4002898).

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

Supplementary Table S1

References

Maas

, Menon

, Adelson

, et al. Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research. Lancet Neurol, 2017; 16(12):987–1048; doi: 10.1016/S1474-4422(17)30371-X

Coronado

, McGuire

, Sarmiento

, et al. Trends in traumatic brain injury in the US and the public health response: 1995-2009. J Safety Res, 2012; 43(4):299–307; doi: 10.1016/j.jsr.2012.08.011

Maas

, Steyerberg

, Butcher

, et al. Prognostic value of computerized tomography scan characteristics in traumatic brain injury: results from the IMPACT study. J Neurotrauma, 2007; 24(2):303–314; doi: 10.1089/neu.2006.0033

Lannsjö

, Backheden

, Johansson

, et al. Does head CT scan pathology predict outcome after mild traumatic brain injury?. Eur J Neurol, 2013; 20(1):124–129; doi: 10.1111/j.1468-1331.2012.03813.x

Sadowski-Cron

, Schneider

, Senn

, et al. Patients with mild traumatic brain injury: immediate and long-term outcome compared with intra-cranial injuries on CT scan. Brain Inj, 2006; 20(11):1131–1137; doi: 10.1080/02699050600832569

Wiesmann

, Steinmeier

, Magerkurth

, et al. Outcome prediction in traumatic brain injury: comparison of neurological status, CT findings, and blood levels of S100B and GFAP. Acta Neurol Scand, 2010; 121(3):178–185; doi: 10.1111/j.1600-0404.2009.01196.x

Thelin

, Johannesson

, Nelson

, et al. S100B is an important outcome predictor in traumatic brain injury. J Neurotrauma, 2013; 30(7):519–528; doi: 10.1089/neu.2012.2553

Cheng

, Yuan

, Yang

, et al. The prognostic value of serum neuron-specific enolase in traumatic brain injury: systematic review and meta-analysis. PLoS One, 2014; 9(9):e106680; doi: 10.1371/journal.pone.0106680

Strand

, Pedersen

, Svingen

, et al. Serum acylcarnitines and risk of cardiovascular death and acute myocardial infarction in patients with stable angina pectoris. J Am Heart Assoc, 2017; 6(2):e003620; doi: 10.1161/JAHA.116.003620

10.

Kalim

, Clish

, Wenger

, et al. A plasma long-chain acylcarnitine predicts cardiovascular mortality in incident dialysis patients. J Am Heart Assoc, 2013; 2(6):e000542; doi: 10.1161/JAHA.113.000542

11.

Chung

, Chen

, Chuang

, et al. Increased plasma acetylcarnitine in sepsis is associated with multiple organ dysfunction and mortality: a multicenter cohort study. Crit Care Med, 2019; 47(2):210–218; doi: 10.1097/CCM.0000000000003517

12.

Widmer

, Thommen

, Becker

, et al. Association of acyl carnitines and mortality in out-of-hospital-cardiac-arrest patients: results of a prospective observational study. J Crit Care, 2020; 58:20–26; doi: 10.1016/j.jcrc.2020.03.009

13.

Bremer

. Carnitine–metabolism and functions. Physiol Rev, 1983; 63(4):1420–1480; doi: 10.1152/physrev.1983.63.4.1420

14.

Reuter

, Evans

. Carnitine and acylcarnitines. Clin Pharmacokinet, 2012; 51(9):553–572; doi: 10.1007/BF03261931

15.

Lifshitz

, Sullivan

, Hovda

, et al. Mitochondrial damage and dysfunction in traumatic brain injury. Mitochondrion, 2004; 4(5-6):705–713; doi: 10.1016/j.mito.2004.07.021

16.

Hiebert

, Shen

, Thimmesch

, et al. Traumatic brain injury and mitochondrial dysfunction. Am J Med Sci, 2015; 350(2):132–138; doi: 10.1097/MAJ.0000000000000506

17.

Chen

, Chan

, Linnane

, et al. L-Carnitine and extendin-4 improve outcomes following moderate brain contusion injury. Sci Rep, 2018; 8(1):1–16; doi: 10.1038/s41598-018-29430-6

18.

Mahmoodpoor

, Shokouhi

, Hamishehkar

, et al. A pilot trial of L-carnitine in patients with traumatic brain injury: Effects on biomarkers of injury. J Crit Care, 2018; 45:128–132; doi: 10.1016/j.jcrc.2018.01.029

19.

Jung

, Ro

, Park

, et al. Vitamin D deficiency and prognosis after traumatic brain injury with intracranial injury: a multicenter observational study. J Neurotrauma, 2022 Jul 21; Epub ahead of print; doi: 10.1089/neu.2022.0053

20.

Maas

, Stocchetti

, Bullock

. Moderate and severe traumatic brain injury in adults. Lancet Neurol, 2008; 7(8):728–741; doi: 10.1016/S1474-4422(08)70164-9

21.

Stevens

, Hillman

, Worthy

, et al. Assay for free and total carnitine in human plasma using tandem mass spectrometry. Clin Chem, 2000; 46(5):727–729; doi: 10.1093/clinchem/46.5.727

22.

Reuter

, Evans

, Chace

, et al. Determination of the reference range of endogenous plasma carnitines in healthy adults. Ann Clin Biochem, 2008; 45(6):585–592; doi: 10.1258/acb.2008.008045

23.

Flanagan

, Simmons

, Vehige

, et al. Role of carnitine in disease. Nutr Metab (Lond), 2010; 7(1):1–14; doi: 10.1186/1743-7075-7-30

24.

Agoston

, Shutes-David

, Peskind

. Biofluid biomarkers of traumatic brain injury. Brain Inj, 2017; 31(9):1195–1203; doi: 10.1080/02699052.2017.1357836

25.

Jones

, McDonald

, Borum

. Acylcarnitines: role in brain. Prog Lipid Res, 2010; 49(1):61–75; doi: 10.1016/j.plipres.2009.08.004

26.

Hoppel

. The role of carnitine in normal and altered fatty acid metabolism. Am J Kidney Dis, 2003; 41:S4–S12; doi: 10.1016/s0272-6386(03)00112-4

27.

Bergsneider

, Hovda

, Shalmon

, et al. Cerebral hyperglycolysis following severe traumatic brain injury in humans: a positron emission tomography study. J Neurosurg, 1997; 86(2):241–251; doi: 10.3171/jns.1997.86.2.0241

28.

Hovda

, Lee

, Smith

, et al. The neurochemical and metabolic cascade following brain injury: moving from animal models to man. J Neurotrauma, 1995; 12(5):903–906; doi: 10.1089/neu.1995.12.903

29.

Brass

, Hoppel

. Carnitine metabolism in the fasting rat. J Biol Chem, 1978; 253(8):2688–2693; doi: 10.1016/S0021-9258(17)40875-1

30.

Bounes

, Faure

, Rouget

, et al. Plasma free carnitine in severe trauma: Influence of the association with traumatic brain injury. Injury, 2018; 49(3):538–542; doi: 10.1016/j.injury.2017.11.005

31.

Mallah

, Quanico

, Raffo-Romero

, et al. Matrix-assisted laser desorption/ionization-mass spectrometry imaging of lipids in experimental model of traumatic brain injury detecting acylcarnitines as injury related markers. Anal Chem, 2019; 91(18):11879–11887; doi: 10.1021/acs.analchem.9b02633

32.

Zheng

, Zhou

, Feng

, et al. Metabolomics analysis of the hippocampus in a rat model of traumatic brain injury during the acute phase. Brain Behav, 2020; 10(2):e01520; doi: 10.1002/brb3.1520

33.

Jeter

, Hergenroeder

, Ward III

, et al. Human mild traumatic brain injury decreases circulating branched-chain amino acids and their metabolite levels. J Neurotrauma, 2013; 30(8):671–679; doi: 10.1089/neu.2012.2491

34.

Kong

, Ro

, Do Shin

, et al. Epidemiology of traumatic brain injury in Korea from 2011 to 2014: based on three major data sources in Korea. J EMS Med, 2021; doi: 10.3561/6/jemsm.2019.00024

35.

Czeiter

, Mondello

, Kovacs

, et al. Brain injury biomarkers may improve the predictive power of the IMPACT outcome calculator. J Neurotrauma, 2012; 29(9):1770–1778; doi: 10.1089/neu.2011.2127

36.

Diaz-Arrastia

, Wang

, Papa

, et al. Acute biomarkers of traumatic brain injury: relationship between plasma levels of ubiquitin C-terminal hydrolase-L1 and glial fibrillary acidic protein. J Neurotrauma, 2014; 31(1):19–25; doi: 10.1089/neu.2013.3040

37.

Steiber

, Kerner

, Hoppel

. Carnitine: a nutritional, biosynthetic, and functional perspective. Mol Aspects Med, 2004; 25(5-6):455–473; doi: 10.1016/j.mam.2004.06.006

38.

Quanico

, Hauberg-Lotte

, Devaux

, et al. 3D MALDI mass spectrometry imaging reveals specific localization of long-chain acylcarnitines within a 10-day time window of spinal cord injury. Sci Rep, 2018; 8(1):1–11; doi: 10.1038/s41598-018-34518-0

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB