A newly developed risk scoring system for mortality among patients with human immunodeficiency virus (HIV) who had developed toxoplasma encephalitis (TE) allowed for accurate prognostication with an expeditious implementation, according to study results published in BMC Infectious Diseases.

To formulate their scoring system, researchers included medical records of 94 patients with HIV and TE hospitalized at 6 centers between 2013 and 2019. Patients were assessed for mortality 6 weeks following their TE diagnosis. To validate their system, 45 patients admitted to the hospital between October 2019 and March 2020 were recruited.

The retrospective cohort were 74.5% men and 38.3% were aged 40 years or older. The 6-week mortality was 10.6%.

Continue Reading

Patients who died were less likely to present with disorders of consciousness (odds ratio [OR], 7.4; 95% CI, 1.817-30.144; P =.005), high CD4+ T-cell count (OR, 7.982; 95% CI, 1.880-33.890; P =.005), dysphasia (OR, 6.375; 95% CI, 1.608-25.274; P =.008), or memory deficits (OR, 4.5; 95% CI, 1.157-17.500; P =.030).

A total of 7 variables were selected for the scoring system and assigned an arbitrary variable based on the regression coefficient. The scoring system included fever (coefficient, 7.764; P =.031; score, 4), low (<25 cells/mL) CD4+ T-cell count (coefficient, 5.431; P =.011; score, 3), disorders of consciousness (coefficient, 5.065; P =.033; score, 3), dizziness (coefficient, 3.624; P =.099; score, 2), presentation 15 or more days after symptom onset (coefficient, 3.34; P =.048; score, 2), memory deficits (coefficient, 3.066; P =.07; score, 2), and patchy brain lesions (coefficient, 2.006; P =.180; score, 1).

Survivors had higher scores than nonsurvivors, and the investigators determined a score of 9 was the optimal risk cut-off threshold (area under the receiver operating characteristic curve [AUC], 0.976; P <.001; sensitivity, 100.0%; specificity, 86.9%).

The scoring system performed well among the verification cohort (sensitivity, 81.8%; specificity, 94.1%; accuracy, 91.1%).

This study was limited by its study population. It is not clear whether this scoring system may have utility among non-Asian patients.

These data indicated a simple scoring system was able to rapidly and effectively assess mortality risk among patients with HIV who had developed TE.


Li Y, Zeng YM, Liu M, et al. Development of a risk scoring system for prognostication in HIV-related toxoplasma encephalitis. BMC Infect Dis. 2020;20(1):923. doi:10.1186/s12879-020-05651-x