[Is dysuria useful for the diagnosis of infection of the urinary tract?].

OBJECTIVE

To assess the validity of criterion and predictive value of dysuria for the diagnosis of UTI.

DESIGN

Cross-sectional study to assess a symptom.

SETTING

Six general medicine clinics (four urban clinics and two rural clinics) in the 11th Health Area in Madrid.

PATIENTS: The sample consists of 232 patients aged above fourteen who consulted during six consecutive months (116 of them reported having dysuria and 116 were asymptomatic).

MEASUREMENTS AND MAIN

RESULTS

The diagnosis of urinary tract infections (UTI) was achieved through positive urine cultures or bacteriuria and leukocyturia in the centrifuged urine sediment. The sensitivity (S) of dysuria analysis for the diagnosis of UTI, its specificity (E), its predictive value (VP), and its probability coefficient (CP) were considered. Average age of the sample was 54 years old (range 19-82); 73% of the patients were female. No statistically significative difference of sex and age was found between cases and non-cases (p > 0.1). Dysuria showed a 96% of sensitivity (95% CI, 86-98%), a 69% of E (95% CI, 61-76%) and 3.1 of CPP (95% CI, 2.7-3.5) for UTI diagnosis. In the women subgroup there was 95% of sensitivity (95% CI, 84-99%) and 67% of E (95% CI, 58-75%). A positive predictive value of 30% (95% CI, 22-40) and a negative predictive value of 99% (95% CI, 95-100) were estimated for this symptom.

CONCLUSIONS

The diagnosis of urinary tract infections is unlikely in the absence of dysuria, but to treat all dysuria patients as UTI entails a high rate of overtreatment.