PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 4, 2026Journal of Clinical Medicine4 citationsOpen Access

Hypercoagulability in Pulmonary Tuberculosis: Reduced Protein C and Free Protein S Predict Pulmonary Embolism—Evidence from a Prospective Romanian Cohort

View Full Paper
DMDenisa Maria MitroiUniversity of Medicine and Pharmacy of CraiovaSVSilviu Gabriel VlasceanuCarol Davila University of Medicine and PharmacyOZOvidiu Mircea ZlatianClinical Emergency Hospital Bucharest

Key Points

  • This research aims to determine the relationship between protein C, free protein S levels, and pulmonary embolism in active tuberculosis patients.
  • Conducted a prospective cohort study in Romania with 63 patients with active TB and 30 controls.
  • Measured protein C and free protein S levels from venous blood at baseline and 6 months.
  • Evaluated bacillary burden through sputum tests and performed chest CT scans to monitor disease evolution.
  • Utilized multivariable regression and logistic regression for data analysis.
  • TB patients had lower protein C and free protein S levels compared to controls (p < 0.001).
  • Lower anticoagulant levels were associated with higher inflammatory markers and bacillary burden.
  • Reduced protein C and free protein S significantly predicted bacillary persistence (OR ~0.93–0.95 for protein C; ~0.92–0.94 for free protein S).
  • Discrimination for bacillary persistence was high, with AUCs ~0.88–0.89.
  • Patients with pulmonary embolism exhibited lower levels of these anticoagulants than those without.

Abstract

Background/Objectives: Pulmonary tuberculosis (TB) is accompanied by inflammation-driven hypercoagulability and increased venous thromboembolism risk. We investigated whether the natural anticoagulants protein C and free protein S are reduced in active TB and whether baseline levels are associated with bacillary burden, treatment response, CT evolution, and pulmonary embolism (PE). Methods: We conducted a prospective cohort study in Romania, including 63 adults with newly diagnosed, bacteriologically confirmed, drug-susceptible pulmonary TB and 30 TB-free controls (October 2024–December 2025). Venous blood was collected at baseline (before anti-TB therapy) and at 6 months to quantify inflammatory and coagulation parameters, protein C, and free protein S. Sputum AFB smear was assessed at baseline, 2 months, and 6 months; chest CT was performed at baseline and 6 months. Propensity score matching (age, sex, BMI, smoking) and multivariable regression were used to account for confounding. Logistic regression and ROC analyses evaluated the prediction of BK persistence. Results: Compared with controls, TB patients had substantially lower baseline protein C and free protein S levels, and higher D-dimer levels (all p < 0.001). In matched multivariable models, TB status remained independently associated with lower baseline natural anticoagulant levels. Lower baseline protein C and free protein S clustered with higher inflammatory markers and higher bacillary burden, and independently predicted BK persistence at 2 and 6 months (OR per 1%-point increase ~0.93–0.95 for protein C and ~0.92–0.94 for free protein S; all p < 0.001). Discrimination for BK persistence was high (AUCs ~0.88–0.89). Lower baseline levels of natural anticoagulants were also associated with greater residual CT abnormalities at 6 months. PE cases had significantly lower protein C and free protein S than PE-free patients. Conclusions: Active pulmonary TB is associated with marked depletion of protein C and free protein S. Baseline reductions identify patients with higher inflammatory/coagulation activation, higher bacillary burden, delayed microbiological clearance, more residual CT disease, and PE, supporting their potential role as adjunct risk-stratification biomarkers.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Mitroi et al. (2026) studied this question.

synapsesocial.com/papers/69a7cd5ed48f933b5eed9ae8https://doi.org/10.3390/jcm15051903
Ask AI
Helpful
Bookmark
Share
View Full Paper