Asssessing infection before steroids

Epub: Rakusa et al. Testing for urinary tract colonisation before high-dose corticosteroid treatment in acute multiple sclerosis relapses: prospective algorithm validation. Eur J Neurol. 2012. doi: 10.1111/j.1468-1331.2012.03806.x

: To evaluate a dipstick algorithm for urinary tract colonisation, prior to high-dose corticosteroid treatment in acute relapses of MS.

: Prospective cohort study of 267 consecutive MSers with MS relapses requiring corticosteroid treatment in a hospital-based, ambulatory, acute MS relapse clinic. A total of eighteen participants met the exclusion criteria, leaving 249 for analysis. Main outcome measures were urinary dipstick sensitivity, specificity, positive predictive value, negative predictive value and safety of antibiotic co-treatment with high-dose corticosteroids.

: Significant bacteriuria (≥10(5) colonies ml) rate in this population was 11% (95% CI, 7.1-14.9). Specificity and sensitivity of positive leucocyte esterase or nitrite were 78% and 65%. Negative predictive value of urine dipstick was 96%. No clinical adverse events occurred in the 3% (95% CI, 0.9-5.1) of patients with a false-negative dipstick. Eighteen per cent of patients were unnecessarily treated with antibiotics for 48h.

: Urinary dipstick testing allows for rapid and safe management of patients suffering from an acute MS relapse. The algorithm is conservative, and future work is needed to reduce the false-positive rate.


This study illustrates two things the problem with diagnostic tests and the issue of bladder infections and the use of steroids. No diagnostic test is 100% accurate this is why we discuss the issue of sensitivity (the chance of a true positive test being positive) and specificity (the chance of a true negative test being negative). The other metrics we often refer to is the positive and negative predictive value of a test; these vary depending on how common the condition is in the population you are applying the test to. The reason why the dipstick test for urinary infection is not ideal is that it mainly relies on the detection of nitrite in the urine; bacteria make an enzyme called nitrate reductase that converts nitrate to nitrite. The problem with this test is that not all bacteria that commonly cause bladder infections make nitrate reductase; in fact only about 70-80% do."

"It is advisable that if someone is about to receive high-dose steroids for a relapse that we make sure they don't have an infection. Steroids suppress the immune system and put MSers at high-risk of developing septicaemia as a complication of the infection. This is why I personally make sure everybody is infection free before prescribing high-dose steroids."

"As you may be aware that infections often exacerbate preexisting MS symptoms that are often then misinterpreted as being a relapse. This why our contemporary definition of a relapse in clinical trials states that the relapse has to occur in the absence of an infection."