Skip to main content
. 2021 Dec 21;12(1):1. doi: 10.3390/metabo12010001

Table 3.

Metabolomics-assisted prognostication of patients with sepsis non-survivors.

Settings Age (Age Range)
(Sample Size)
APACHE-II
Score
Samples Since Admission Methods Major Findings in Non-Survivors Interpretation Citation
Serum Plasma Blood Metabolic Pathways Decreased Increased
48-H mortality 67 ± 15 (9)
vs.
63 ± 18 (26)
26 ± 6
vs.
20 ± 8

within 48-H before death
Targeted
(LC-MS/MS)
Amino acids and amines -
  • S-(3-methyl-butanoyl)dihydro-lipoamide-E

Amino acids and phospholipids could indicate the possibility of death within 48-H in patients with sepsis [10]
Phospholipids -
  • Phosphatidyl-glycerol(22:2

  • (13Z,16Z)/0:0)

  • GPC

7-D mortality 60 (36–80)
(9)
vs.
60 (27–84)
(13)
31 (16–46)
vs.
22 (14–38)

within 48 h = D1, D3, and D7
Targeted
(LC-MS/MS)
Persisted D1 to D7 Fatty acids and proresolving lipids signal 7-D mortality in critically-ill patients with sepsis [60]
Fatty acids -
  • Prostaglandin F2α

  • Leukotriene B4

  • Resolvin E1

  • Resolvin D5

  • 17R-Protectin D1

At D1
Fatty acids -
  • 7-HDHA

  • 17-HDHA

  • 15-HEPE

  • 18-HEPE

  • 15-HETE

-
  • Protectin D1

At D3
Fatty acids -
  • 17-HDHA

  • 18-HEPE

  • 5-HETE

  • 15-HETE

  • 5S,12S-diHETE

-
  • 17-epi-Resolvin D1

  • 17-epi-Protectin D1

At D7
Fatty acids -
  • 4S,14S-diHDHA

  • 5S,15S-diHETE

-
  • Resolvin E2

28-D mortality 69 ± 17
(31)
vs.
56 ± 19
(90)
23 ± 8
vs.
15 ± 7

H0
and H24

H0
and H24
- Untargeted (UPLC-MS/MS
and
GC-MS)
- Targeted (UPLC-MS/MS)
Persisted At H0 to H24 28-D mortality could be predicted by several amino acids, amines, fatty acids, and glycolysis metabolites [61]
Amino acids and amines -
  • N-Acetylthreonine

  • 1-Methylimidazole acetate

  • N-Acetylalanine

  • Hydroxyproline

  • Prolylhydroxy-proline

  • SDMA

  • AC

  • C3, C4, C5, C5:1

Fatty acids -
  • AC

  • C6, C8, C10, C16, C18

  • 3-Hydroxy-2-ethyl-propionate

Glycolysis -
  • Erythronate

  • N-acetylneura-minate

  • Arbitol

  • AC C2

TCA cycle -
  • Malate

At H0
Fatty acids -
  • AC C12

At H24
Amino acids and amines -
  • α-Hydroxyiso-valerate

  • Glutaroylcarnitine

  • Hydroxyisovalero-ylcarnitine

Fatty acids -
  • Hexadecanedioate

Phospholipids
  • 1-Arachido-nyl-GPE

  • 1-Palmitoyl-GPC

  • 1-Stearoyl-GPC

  • 1-Eicosatri-enoyl-GPC

  • 1-Arachido-noyl-GPC

-
28-D mortality T: 58 ± 15
(30)
vs.
53 ± 14
(60)
V: 69 ± 16
(34)
vs.
58 ± 17
(115)
30 ± 11
vs.
23 ± 9
23 ± 8
vs.
15 ± 7

H0
Targeted
(GC-MS and
LC-MS)
Amino acids and amines -
  • Ornithine

  • Kynurenine

  • AC C3, C4, C5, C5-OH, C5:1

  • β-hydroxyiso-valerate

  • N-acetylalanine

  • N-acetylserine

  • α-Hydroxy-isovalerate

  • γ-glutamylphenyl-alanine

  • γ-glutamyltyrosine

Non-surviving 28-D sepsis patients had specific changes in amino acids, fatty acids, glycolysis, and bile acids’ metabolic pathways, as well as an increase in aromatic microbial metabolites [62]
Fatty acids -
  • AC C6

Phospholipids
  • 1-arachidonyl-GPE (20:4)

  • 1-arachidonyl-GPC (20:4)

  • 1-linoleoyl-GPC (18:2)

  • 2-palmitoyl-GPC (16:0)

  • 1-palmitoyl-GPC (16:0)

  • 1-stearoyl-GPC (18:0)

-
Glycolysis
Aromatic microbial metabolites
-

-
  • Sucrose

  • Lactate

  • 3-(4-hydroxyphenyl) lactic acid

28-D mortality 61 ± 21
(15)
vs.
54 ± 23
(20)
22 ± 8
vs.
10 ± 5

within 24H = D1, D3, D5, D7, D10, and D14
Targeted
(LC-MS/MS)
At certain time points Amino acids could indicate the possibility of death in septic patients [11]
Amino acids and amines
  • Arginine

  • Glutamate

  • Serine

  • Phosphoserine

  • Taurine

  • Tryptophan

  • α-Aminoadipic acid

  • Cystathionine

  • Ethanolamine

  • Phenylalanine

28-D mortality 68 (51–75)
(31)
vs.
63 (53–74)
(89)
12 (8–9)
vs.
9 (6–13)

within
24H = D1, D3, D7
Targeted
(LC-MS/MS)
Amino acids and amines
  • SDMA

  • ADMA

High level of plasma SDMA and ADMA can predict sepsis non-survival [66]
28-D mortality 70 ± 13
(21)
vs.
72 ± 15
(69)
26 ± 9
vs.
23 ± 8

H0
Targeted
(UHPLC-MS)
Glycolysis -
  • AC C2

Acetylcarnitine can forecast 28-D mortality in patients with sepsis [63]
28-D mortality 67 ± 14
(54)
vs.
62 ± 19
(134)
22 (18–30)
vs.
18 (13–24)

H0
Targeted (LC-MS) Amino acids and amines -
  • Isoleucine

  • Alanine

Particular metabolites can forecast 28-D mortality in sepsis patients [64]
Phospholipids
  • LysoPC C22:0

  • LysoPC C24:0

-
Glycolysis -
  • Lactate

  • Pyruvate

  • AC C2

30-D mortality 55 (17–80)
(39)
vs.
54 (20–91)
(63)
N/A
vs.
N/A

D1, D4,
and D11
Targeted
(LC-MS/MS)
Lipids
Persisted along D1 to D11 [8]
Fatty acids and phospholipids -
  • Total ceramides-to-SM ratio/total LysoPC-to-PC ratio

90-D mortality 75 ± 13
(30)
vs.
71 ± 13
(63)
9 ± 4 $
vs.
5 ± 4 $

D1
Targeted
(UPLC-MS)
Amino acids and amine -
  • Phenylalanine

  • Leucine

In sepsis patients, 90-D mortality can be expected by phenylalanine and leucine [65]

Continuous data are presented in mean ± SD, otherwise reported as median and IQR1-3. $ Sequential Organ Failure Assessment (SOFA) score. Abbreviations: 7-HOCA, 7-α-hydroxy-3-oxo-4-cholestenoate; APACHE-II score, Acute Physiology and Chronic Health Evaluation-II score; ADMA, asymmetric dimethylarginine; C, number of carbons in the fatty acid side chain; D, Day; GPC, Glycerophosphocholine; GPE, Glycerophosphoethanolamine; H, Hour; HDHA, Hydroxydocosahexaenoate; HEPE; Hydroxyeicosapentaenoate; HETE; Hydroxyeicosatetraenoate; LysoPC, lysophosphatidylcholine; LC-MS/MS, liquid chromatography-tandem mass spectrometry; SDMA, symmetric dimethylarginine; UHPLC-MS, Ultra-high performance liquid chromatography mass spectrometry; UPLC-MS, Ultra-performance liquid chromatography-mass spectrometry; T, training cohort; V, validation cohort.