Serum c-reactive protein as a biomarker in differentiating cryptococcal iris and cryptococcal relapse among HIV patients in Mulago hospital

Abstract

Introduction: The mortality after Cryptococcal Meningitis is still high, in part due to Immune Reconstitution Inflammatory Syndrome (IRIS) related events. Cryptococcal Meningitis paradoxical IRIS incidence rates of 45% with a 36% mortality rate have been noted in a Mulago Cohort. Cryptococcal relapse rates of 23% have been noted in South Africa. Clinically it is difficult to differentiate paradoxical cryptococcal IRIS and cryptococcal relapse. CSF Cultures are required yet they are not readily available and have a long turn around time. It is important to differentiate these conditions because the therapeutic approach differs. Serum C-Reactive Protein (CRP), an inflammatory marker is elevated in cryptococcal menigitis and has been shown to predict paradoxical CM IRIS, hence the need to perform this study to determine whether it distinguishes paradoxical CM IRIS and CM relapse. The objectives of the study were to compare the serum CRP level in patients with paradoxical Cryptococcal Meningitis IRIS and in those with cryptococcal meningitis relapse and to compare the serum CRP level in patients with ART associated cryptococcal meningitis and ART naive patients presenting with cryptococcal meningitis. Methods This was a cross-sectional comparative study, with a descriptive component. We used information and stored serum samples from patients who developed CM paradoxical IRIS in the concluded Cryptococcal Optimal ART Timing (COAT) Trial and recruited CM relapse patients from the on going pilot study for Adjunctive Sertraline for the treatment of HIV associated cryptococcal meningitis (ASTRO) trial in the period between September 2014-March 2015. Demographic, clinical and laboratory analysis with CSF confirmation of cryptococcal infection was done in all the patients. Comparisons of serum CRP for the paradoxical CM IRIS and CM relapse patients was done in samples collected on the day of enrollment. Results A total of 124 patients including 33 paradoxical CM IRISpatients from the COAT study, 29 CM relapse patients from the ASTRO study, 31 ART-associated CM patients and 31 ART naive patients with CM were recruited. The median (IQR) serum CRP in patients with paradoxical CM IRIS, 51.6mg/L (IQR 24.7-101) was higher, but not statistically significant when compared to CM relapse patients, 36.4mg/L (IQR 32.6-57, p = 0.157). The serum CRP in ART naive patients with CM, 53.5mg/L (IQR 21.7-133.4), was higher but not stastistically significant when compared to ART-associated CM patients, 30.1mg/L(31.2-77.1, p = 0.310) There was no difference in the CSF white cell counts of the CM relapse and paradoxical IRIS group. Conclusion Serum CRP does not distinguish paradoxical CM IRIS and CM relapse. The gold standard to distinguish this clinical phenomena remains CSF cultures. Like wise, CSF white cell counts did not distinguish cryptococcal paradoxical IRIS and relapse. There is no statistically significant difference in the serum CRP and CSF white cell count between the ART naive and ART-associated CM patients.