Many scientists are currently attempting to identify a urinary biomarker for Overactive Bladder Syndrome (OAB). In these studies, urine creatinine is commonly used to normalise for urinary proteins. In the field of urinary cytokines, some authors routinely perform creatinine normalisation [1] while others do not [2]. However, urinary creatinine concentrations can vary widely, leading to differing normalised results. As such, there has been an increasing trend to use urine osmolarity (as specific gravity) as a normalising factor. The present study aims to compare urinary creatinine and specific gravity with patient age, body mass index (BMI) and clinical history, to determine which is the better normalising factor.

As part of an ongoing study of urinary cytokines in OAB patients, midstream urine (MSU) samples were collected from women (median age of 64 (52-72)) with urinary incontinence at an urogynaecology clinic. A standardised history was routinely taken. The presence of leukocytes and specific gravity were determined by dipstick. Creatinine concentrations were quantitated by a creatinine ELISA (Sigma). Presence of a current UTI was assessed by the microbiology department (>107 CFU/mL). Data are presented as median and interquartile range (IQR); analysis by linear regression and Mann-Whitney t-tests.

Specific gravity (n=168 samples) ranged from 1.0 to 1.03. Patients with hypercholesterolemia had a significantly greater specific gravity (p<0.01, Table 1). Current or history of UTI did not affect specific gravity, however the presence of pyuria did (p<0.05, Table 1). There was a significant correlation of specific gravity and BMI (p<0.05) but not with age (Figure 1A). Urine creatinine was more highly variable, ranging from 0.25 to 46.5 µg/mL (n=136 samples). The normal range for urinary creatinine concentrations is 3-30µg/mL (WHO Guidelines [3]). However, 32% (44/136) of samples were found to be outside of this range. Of these, 11% (5/44) were <3µg/mL while 89% (39/44) were >30µg/mL. There was no significant difference in the urinary creatinine concentration with either patient age or BMI (Figure 1B), nor in relation to a variety of clinical conditions (Table 1). In addition, urinary creatinine was found to be inversely correlated with specific gravity (r=-0.30, p=0.0003).

Urinary creatinine has been used previously as a normalisation factor for studies of urinary proteins such as cytokines. However, our study found that the concentration of urinary creatinine was highly variable, with 32% of samples found to be outside the normal range. It has been previously suggested that urinary creatinine concentration decreased with age [3], however this correlation was not found in the current study. Our results also did not show any significant variation in urinary creatinine with the clinical conditions examined. 
Somewhat surprisingly there was an inverse correlation between urinary creatinine and urine osmolarity. Urine osmolarity (as specific gravity) was found to be less variable than urinary creatinine in the population studied. Specific gravity was influenced by patient BMI, the presence of hypercholesterolemia and the presence of pyuria.

The results of this study do not clearly demonstrate the superiority of either urinary creatinine or specific gravity as a normalisation factor in this OAB patient sample. While the results for creatinine were not influenced by comorbidity, the wide variability in urinary concentrations of creatinine causes concern, as does the lack of correlation with age. Specific gravity, on the other hand, showed less variability but was affected by conditions that are common in OAB patients.

Tyagi et al., 2010 Urine cytokines suggest an inflammatory response in the overactive bladder: a pilot study. Int Urol Nephrol 42: 629-635
Kim et al., 2006 Nerve growth factor and prostaglandins in the urine of female patients with Overactive Bladder, J Urol 175(5): 1773–1776
Barr et al (2005) Urinary creatinine concentrations in the U.S. population: Implications for urinary biologic monitoring measurements. Env Health Persp 113(2):192-200.