Updated Algorithm Improves Specificity of UTI Detection in Patients With Relapsing Multiple Sclerosis

With an improved specificity, an updated algorithm for the detection of urinary tract infections (UTIs) in patients with relapsing multiple sclerosis (MS) addresses previous issues concerning the unnecessary prescription of antibiotics, whilst improving accuracy and maintaining safety. This according to a study published in the Journal of Neurological Sciences.

Urinary tract infections have been linked to worsening of neurological symptoms, and implicated as triggers of acute MS relapses. Ruling out UTI prior to high dose corticosteroid treatment treatment is both critically important and clinically challenging. In a previous algorithm, a urinary dipstick was performed on all patients with a relapse and those with positive urinary nitrites or leucocyte esterase received empirical antibiotics along with high dose corticosteroids. This algorithm was found to have a high false positive rate, with patients being treated with antibiotics unnecessarily. The goal of this study was to reduce the false positive rate found in the previous algorithm while maintaining accuracy, safety and efficiency in patients suffering from MS relapse.

The study consisted of a prospective cohort of consecutive patients with MS relapse presenting to an MS clinic in London between 2014 and 2017 (n=299). A previous algorithm deemed a high likelihood of UTI in patients who tested positive for either urinary nitrites or leucocyte esterase. In this updated algorithm, patients were deemed to have a UTI, and subsequently treated with high-dose corticosteroids, after testing positive for 2 or more of the following: nitrites, leukocyte esterase, and/or cloudy urine. Individuals with clinically confirmed relapse of MS and a UTI were treated with high-dose intravenous methylprednisolone (1 g) for 3 days or oral methylprednisolone (0.5 g) for 5 days.

A total of 33 patients (11%; 95% CI, 8%-15%) presented with significant bacteriuria (>10⁵ colonies/mL). The updated algorithm was found to have a sensitivity of 24% and a specificity of 94%. The

positive and negative predictive values were 35% and 91%, respectively. Overall, the algorithm’s accuracy (per-treatment) was found to be 87%. In the 25 patients who received high-dose methylprednisolone and had an untreated UTI, no adverse sequelae was reported.

Limitations of the study included the small sample size as well as the recruitment of patients from a single center.

The researchers suggest that this new algorithm “allows patients with significant relapses to be treated quickly with [high-dose corticosteroids] in an outpatient setting, while providing a mechanism to reduce unnecessary antibiotic exposure.”

Reference

O’Herlihy F, John NA, Li V, et al. Screening for urinary tract colonisation prior to corticosteroid administration in acute multiple sclerosis relapses: validation of an updated algorithm [published online September 16, 2019]. J Neurol Sci. doi:10.1016/j.jns.2019.116456