Introduction: Phenylalanine, tyrosine, and tryptophan metabolites have various diagnostic and prognostic significance for the development of infectious complications in post-surgery and critically ill patients. Historically, protocols based on gas chromatography–mass spectrometry (GC-MS) with liquid–liquid extraction (LLE) [DOI:10.1134/s1061934818020089] and microextraction by packed sorbent (MEPS) [DOI:10.3390/molecules25143258] have been used in clinical studies. However, recently developed and validated protocols using ultra-high-pressure liquid chromatography–tandem mass spectrometry (UPLC-MS/MS) [DOI:10.3390/metabo13111128; DOI:10.1016/j.jpba.2025.116803] have advantages and potential for more active clinical application.

The aim of this study was to compare the analytical characteristics of the aromatic metabolite determination in the blood serum and cerebrospinal fluid (CSF) using GC-MS and UPLC-MS/MS.

Methods: LLE or MEPS sample preparation, drying, and silylation were performed before GC-MS analysis of 8 phenyl- and 5 indole-containing metabolites using Trace GC 1310 ISQ LT. We used the protocol involving protein precipitation, or the protocol with additional evaporation and the dissolution of dry residue, and UPLC-MS/MS analysis of 6 phenyl- and 5 indole-containing metabolites, conducted using Water Acquity UPLC and AB Sciex QTRAP 5500. Serum samples from 48 healthy donors and CSF samples from 133 post-neurosurgical patients were analyzed.

Results: LLE with GC-MS resulted in aromatic metabolite determination in the serum and CSF at 0.5-40 and 0.09-27 μmol/L linear concentration ranges, respectively. MEPS with GC-MS resulted in a 0.4-18 μmol/L linear concentration range for serum and CSF. Protein precipitation with UPLC-MS/MS resulted in a 0.02-25 μmol/L concentration range for serum. The protocol with additional steps resulted ina 0.2-375 nmol/L concentration range for CSF. GC-MS and UPLC-MS/MS serum analysis of healthy donors resulted in the complete comparability of the 4-hydroxyphenyllactic acid concentrations. However, UPLC-MS/MS CSF analysis of post-neurosurgical patients resulted in better sensitivity than GC-MS.

Conclusions: GC-MS and UPLC-MS/MS can be used for clinical analysis of serum samples, while UPLC-MS/MS is preferred for CSF analysis.