Rapid Responses to:

PRIMARY CARE: Dee Richards, Les Toop, Stephen Chambers, and Lynn Fletcher Response to antibiotics of women with symptoms of urinary tract infection but negative dipstick urine test results: double blind randomised controlled trial BMJ 2005; 0: bmj.38496.452581.8Fv1 [Abstract]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Another way to treat dysuria

Uffe Ravnskov   (21 July 2005)

Diagnosis of urinary infection

Philip Toozs-Hobson, James Gray, Arri Coomarasamy   (9 August 2005)

Authors reply: Diagnosis of urinary infection

Dee A Richards, Les Toop, Stephen Chambers, Lynn Fletcher   (15 August 2005)

Another way to treat dysuria 21 July 2005
Uffe Ravnskov, MD, PhD, independent researcher Magle Stora Kyrkogata 9, 22350 Lund, Sweden Send response to journal: Re: Another way to treat dysuria	In my experience the commonest cause of dysuria is soap. Therefore it seems bad advice to treat this disorder with antibiotics without further investigation, as suggested by Dr. Richards and coworkers.1 In a prospective study of women, who consulted me because of dysuria and/or frequency I found that all of 14 women with the urethral syndrome (dysuria without bacteriuria), 15/17 with uncomplicated, lower urinary tract infection (dysuria with bacteriuri), but only 6/19 with asymptomatic bacteriuria used soap or other detergents on the sexual organs regularly.2      All women were advised to use water only and women with a positive urine culture were treated with trimetoprim for three days. At follow-up 22 of the 31 women with dysuria had stopped or substantially reduced their use of soap. In seventeen dysuria had disappeared completely after 1-8 weeks. In contrast, six out of seven who did not follow my advice had still dysuria.      Interestingly, recurrences of bacteriuria were asymptomatic. Evidently, soap washing of the outer genitals rather than bacteria seems to be the primary cause of dysuria. As there is no evidence that asymptomatic bacteriuria in non-pregnant women has any adverse health effects, a better treatment of dysuria, whether bacteriuria is present or not, is to ask the women to wash with water only. Richards D, Toop L, Chambers S, Fletcher L. Response to antibiotics of women with symptoms of urinary tract infection but negative dipstick urine test results: double blind randomised controlled trial. BMJ 2005;331:143-147. Ravnskov U. Soap is the major cause of dysuria. Lancet 1984;2:1027-8. Competing interests: None declared
Diagnosis of urinary infection 9 August 2005
Philip Toozs-Hobson, Consultant Urogynaecologist Birmingham Women's Hospital, B15 2TG, James Gray, Arri Coomarasamy Send response to journal: Re: Diagnosis of urinary infection	Sir, We read with interest the article by Richards et al (1) which raises some interesting points on the diagnosis and treatment of Uropathogens. The concept of significant bacteriuria was developed by Kass and colleagues in the mid-1950s, on the basis that quantitative culture could help distinguish between the presence of bacteria multiplying in the urine and bacteria introduced as contaminants from the urethra or introitus during voiding of a midstream sample of urine. Based on the fact that most urinary tract infections are caused by Escherichia coli and related Gram- negative bacteria that multiply rapidly in urine, significant bacteriuria was defined as a microbial count of greater than 100 000 colony-forming units (cfu) per ml (2), although even using this criterion a single culture has up to a 20% chance of representing contamination only (3). Following the widespread adoption of quantitative urine cultures, it was noted that 20% to 40% of women with symptoms of acute urinary tract infection (UTI) have bacterial counts of less than 100 000 cfu per ml. Further evidence that UTI can be associated with bacterial counts of less than 100 000 cfu per ml stems from the observations that the species of bacteria isolated from these women are the same as those from women with higher bacterial counts, bacteria can be isolated from urine samples collected directly from the bladder (e.g. by catheterization or suprapubic aspiration), and symptoms often respond to antimicrobial therapy. Some studies have reported that bacterial counts as low as 100 cfu per ml can be associated with symptoms. However, up to 10% of asymptomatic women have bacterial counts of this magnitude in urine. As the bacterial count increases within the range 100 to 100 000 cfu per ml the greater the association with symptoms and in the magnitude of pyuria. Thus there is no reliable cut-off for bacteriuria to be considered significant, but the importance of bacterial counts of less than 100 000 cfu per ml should be assessed in the light of symptoms and pyuria. There are various possible explanations for the finding of low level bacteriuruia in symptomatic UTI, including concurrent use of antimicrobials or that it represents an early phase of the infection. We note that 40% of the samples whilst having a negative dipstick had pyuria on microscopy. Although microscopic findings did not predict response to treatment, it is unlikely that the study was adequately powered to verify this. To our reading these data demonstrate that a negative dipstick does not reliably exclude UTI in symptomatic women. The results of this study may be partly explained by the infection being in an early phase, and therefore associated with low level bacteruria and pyuria prior to an imflammatory response. To these ends data on the duration of symptoms prior to presentation would be helpful in interpreting the results before we throw the baby out with the bathwater. Yours Faithfully Philip Toozs-Hobson Consultant Urogynaecologist James Gray Consultant Microbiologist Arri Coomarasamy Specialist Registrar Birmingham Women’s Hospital, Birmingham UK 1. Response to antibiotics of women with symptoms of urinary tract infection but negative dipstick urine test results:double blind randomised controlled trial Dee Richards, Les Toop, Stephen Chambers, Lynn Fletcher. BMJ 2005;331, 143-146 2. Kass EH. Asypmtomatic infections of the urinary tract. Trans Assoc Am Phys. 1956; 69: 56-64. 3. Kass EH. The role of asypmtomatic bacteriuria in the pathogenesis of pyelonephritis. In Quinn EL, Kass EH (eds). Biology of Pyelonephritis. Boston: Little, Brown 1960: 399-406. Competing interests: None declared
Authors reply: Diagnosis of urinary infection 15 August 2005
Dee A Richards, senior lecturer Department of Public Health and General Practice, Christchurch School of Medicine and Health Science, Les Toop, Stephen Chambers, Lynn Fletcher Send response to journal: Re: Authors reply: Diagnosis of urinary infection	We would like to respond to the points raised by Philip Toozs-Hobson and colleagues regarding our RCT of antibiotic treatment of women with symptoms of urinary tract infection but negative dipstick urine test results. Low count bacteriuria/pyuria and response to treatment: Only 6 participants had low count bacteriuria and they were evenly distributed between the placebo and treatment groups. Of those in the treatment group only one responded. We calculated the treatment effect removing those women with pyuria and the effect persisted and was still significant (ƒÓ2 =7.82 df =1, p=0.005). We also removed all those with low count bacteriuria or pyuria and calculation showed that the effect persisted and was still significant (ƒÓ2 =8.33, df =1, p=0.004). Early stage UTI: We assessed, independently of low count bacteriuria on initial MSU, the possibility of this group of women simply representing "early but ordinary" UTI. We believe this is unlikely for two reasons. We collected a second MSU specimen on all women at day 7. Only 3 women in the placebo group grew a uropathogen in their MSU at 7 days suggesting that this group were not simply in the early stages of UTI development. We also assessed duration of dysuria prior to treatment in the dipstick negative group The duration of dysuria (median four days) was in fact longer than in a group of dipstick positive women recruited to another study simultaneously (median 2 days). We agree that infection is the most likely cause of symptoms - presumably either with very low numbers of conventional organisms or with organisms not being recognised. However these data suggest that neither dipstick testing nor routine laboratory testing are detecting this infection. We therefore believe that negative dipstick and negative urine findings cannot be relied on as intermediate predictors of those unlikely to benefit from treatment. Until we have better methods of predicting who will respond to antibiotics, empirical treatment of all with symptoms of uncomplicated UTI seems appropriate. Dee Richards Les Toop Stephen Chambers Lynn Fletcher Competing interests: None declared