The aims of this study are to examine voiding patterns in patients with nocturnal polyuria (NP) and identify a novel predictive indicator based on voiding frequency ratios.

Patients who presented with a chief complaint of nocturia at 17 institutions between January 2018 and December 2022 were retrospectively reviewed. Initially, 909 patients were included in the study, and after applying exclusion criteria (e.g., bladder stones, urinary tract infections, genitourinary malignancies, or current desmopressin use), 875 patients were included in the final analysis. Data collected included demographic information (age, sex, body mass index, and blood pressure), medical history, blood test results (including glomerular filtration rate), Eastern Cooperative Oncology Group Performance Status, residual urine volume, current medications, Nocturia Quality of Life Questionnaire scores, Overactive Bladder Symptom Score, and bladder diary data. Prostate-specific antigen levels, prostate volume, and the International Prostate Symptom Score were recorded only for male patients. Patients were classified into NP (NPi: nocturnal polyuria index > 33%) and non-NP (NPi ≤ 33%) groups. Daytime and nighttime voiding frequencies and volumes were collected from bladder diaries. The correlation between the NPi and voiding parameters was assessed using Spearman's rank correlation. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the predictive ability of the nighttime-to-24-hour voiding frequency ratio for NP.

Of the 875 participants, 655 (74.9%) were male and 220 (25.1%) were female. The median (IQR) age of all participants was 75 (69-80). The median of the daytime voiding frequency was significantly less in the patients with NP (7.0, IQR 6.0-9.0) compared to the patients without NP (8.7, IQR 7.3-10.4, p<0.001). The median of the daytime voiding volume was significantly lower in the patients with NP (146ml, IQR 107-191) compared to the patients without NP (153ml, IQR 118-194, p=0.028). ROC curves were generated, and the AUCs with and without NP were calculated for the parameters showing the ratio of nighttime to 24-hour voiding frequency and NPi. The optimal cut-off value was determined using the Youden index. For the ratio of nocturnal to 24-hour voiding frequency, the AUC predicting NP was 0.79 (p < 0.001) for all patients, 0.80 (p < 0.001) for men, and 0.76 (p < 0.001) for women. The optimal cut-off values were 0.23 (sensitivity: 75%, specificity: 68%) for all patients, 0.23 (sensitivity: 77%, specificity: 68%) for men, and 0.23 (sensitivity: 71%, specificity: 71%) for women.

Patients with NP urinated significantly less frequently and produced more urine during the day than those without NP (p < 0.001). The ratio of nocturnal to 24-hour voiding frequency showed a significant positive correlation with NPi (r = 0.38, p < 0.001) and was the strongest predictor of NP. ROC curve analysis showed that this ratio could predict NP, with an area under the curve (AUC) of 0.79. A cutoff value of 0.23 provided a sensitivity of 75% and a specificity of 68%.

The results of the present study validated the diurnal urinary pattern in patients with NP and showed that the urinary frequency ratio may be predictive of NP. Patients with NP demonstrated significantly lower daytime voiding frequency and urine volume than those without NP. The ratio of nighttime to 24-hour voiding frequency provided a rough predictive ability for NP at initial consultation (sensitivity: 75%, specificity: 68%, cutoff: 0.23). With additional data and further research, this indicator has the potential to become more practical and accurate.

Koji Iinuma et al (2023) The prevalence and predictive factors of nocturnal polyuria in Japanese patients with nocturia: a multicentral retrospective cohort study. Scientific Reports. 13:18128. https://doi.org/10.1038/s41598-023-45311-z
Nishiwaki T et al (2025) Prediction of nocturnal polyuria in patients with nocturia based on voiding frequency. Int Urol Nephrol. Epub ahead of print. 10.1007/s11255-025-04402-8.