Assessment of the Most Impactful Combination of Factors Associated with Nocturia and to Define Nocturnal Polyuria by Multivariate Modelling.

Nocturia is common and associated with multiple disease states. Many potential mechanisms have been proposed for nocturia, which also remains challenging to manage.

To use multivariate analysis to determine which combinations of factors can accurately discriminate clinically significant nocturia in patients to facilitate clinical management and treatment decisions.

Data analysis was based on frequency volume charts from three randomized controlled trials. There were 1479 patients included, of which 215 patients had no/mild nocturia and 1264 had clinically significant nocturia with at least two voids per night. Factors studied that may influence nocturia were demographics, sleep duration, functional bladder capacity, 24 h urine volume and literature-suggested definitions of nocturnal polyuria. We used univariate analysis and cross-validated multivariate modelling to assess association between factors and nocturia status, redundancy between factors and whether the combined use of factors could explain patients' nocturia status.

The multivariate analyses showed that the most useful definitions of nocturia are 'Nocturia Index' (NI) and 'Nocturnal Urine Production per hour' (NUPh) in combination with functional bladder capacity and sleep duration. Published definitions providing binary nocturnal polyuria outcomes had lower performance than continuous indices. These analyses also showed that NI was not specific to nocturnal polyuria as it also captured nocturia due to low functional bladder capacity. By contrast, NUPh was demonstrated to be specific to nocturnal polyuria.

NUPh has previously been shown among elderly males to be essential in nocturia and a very valid measure of nocturnal polyuria. However, the current, large and independent dataset now confirms that it can be applied in an adult population with a complaint of nocturia covering both males and females.

Journal of clinical medicine. 2020 Jul 16*** epublish ***

Tine Olesen, Jerome Paul, Pierre Gramme, Marcus J Drake, Johan Vandewalle, Karel Everaert

Urology Department, Ghent University Hospital, 9000 Ghent, Belgium., DNAlytics, 1348 Louvain-la-Neuve, Belgium., Bristol Urological Institute, University of Bristol, Bristol BS105NB, UK., Department of Pediatric Nephrology, Safepedrug, University Hospital Ghent, 9000 Ghent, Belgium.