Bladder cancer is a primary cancer of the urothelium, with relatively high incidence and mortality worldwide (1). Muscle-invasive bladder cancer (MIBC) accounts for about 25% of the bladder cancer cases (2). Because of the high risk of recurrence and metastasis, radical cystectomy (RC) and urinary diversion is often recommended when diagnosed at the advanced stage.

In the recent decades, despite the improvement of surgical techniques and advances in preoperative management, RC and urinary diversion have still been acknowledged to be the most complicated surgery in urology (3). The associated risk of RC and urinary diversion is based not only on the technical challenges of the surgery, but also on the characteristics of the patients. Patients who underwent RC and urinary diversion are often of advanced age and accompanied by comorbidity and malnutrition. Many studies have reported common complications after RC such as incisional infection, mechanical bowel obstruction, bleeding, lymphatic leak, and ureteroenteric anastomotic stricture (4–6). The incidence of early complications after RC (within 90 days after surgery) is reported to be 20–57% (7, 8) and postoperative infection is one of the most common complications, especially urinary tract infection (UTI) (9). Considering a substitution of intestine for urinary tract reconstruction, patients who underwent urinary diversion are particularly susceptible to UTI.

To predict the common complication of UTI for an early management, researchers (6, 10–15) have made efforts to find possible risk factors and microbiology of UTI within 90 days after RC and urinary diversion. The postoperative 90-day UTI rates vary across different patient selection and urinary diversion types. Early UTI increases the readmission rate and hospitalization cost. Therefore, it is critical to predicting the probability of early UTI.

In this study, we conducted a retrospective study to analyze the risk factors of UTI within 90 days in the patients with bladder cancer after RC and urinary diversion. Moreover, the microbiology of UTI was also identified. To our knowledge, this study is the first attempt to develop and validate a predictive nomogram for this particular population to determine the possibility of incidence of UTI within 90 days after RC and urinary diversion.

Postoperative UTI is very common in patients who underwent RC and urinary. Several inherent risk factors could account for the postoperative infectious complications. One reason is that urinary diversion operation often involves the intestine. The urinary tract reconstruction often attaches part of the intestine to the upper urinary tract, especially in patients undergoing IC and ONB. The second reason is the colonization of bacteria caused by the ureteral stent indwelling. Accurate predictions about the UTI events after RC and urinary diversion are important for timely intervention. Thus, a predictive nomogram was developed, and included risk factors that were easy to acquire before surgery. Importantly, the performance of our model was also supported by the C-index and the calibration curve with the great discriminatory ability and high-predictive accuracy.

The UTI rate was relatively high in this study compared with the previous reports (27.3 vs. 9–10.3%) (6, 10, 11, 13, 15). One potential reason for this is the different definitions we used for UTI. In the aforementioned reports, UTI is defined as positive urine culture with clinical symptoms, while in our study, patients with asymptomatic bacteriuria were also included. Generally, a positive urine culture alone would not be regarded as clinically significant without accompanying fever or clinical complaints. However, patients who underwent urinary diversion with a positive urine culture are considered to be complicated UTI according to the EAU guideline. For complicated UTI, clinical treatment is difficult. It is more likely to progress to systemic and severe infection. As a result, all urologists should pay more attention to such infections and use antibiotics rationally.

With respect to the distribution of microbiology, most studies (11, 13, 14, 16) reported a predominance of Gram-negative bacteria. However, our result clearly showed a high rate of Gram-positive species in isolated cultures. In addition, we noted that fungi account for a relatively high rate (13.9%). Parker et al. (10) similarly report high rates of Enterococcal species (34.5%), Staphylococcus aureus (17.7%), as well as Escherichia coli (17.7%). With the application of antibiotics worldwide, the bacterial spectrum may vary over time. Notably, the frequency of antibiotic prophylaxis usage may lead to particular resistance of bacteria. In this study, the detailed analysis of the microbiology and antibiotic resistance highlights the importance of broadening antibiotic prophylaxis for the patients undergoing RC and urinary diversion. Meanwhile, it cannot be ignored of antimycotic medication in this particular population.

In this study, the univariate and subsequent multivariate logistic analyses determined that CCI, PNI, stricture, and urinary diversion type are predictors of UTI. Therefore, a nomogram is presented concisely with these factors. Other studies have reported different risk factors including diabetes (10), BMI (17, 18), female gender (19), urine leakage, and receipt of a perioperative blood transfusion (10). Kim et al. (16) indicated that only ureteral stricture was independent risk factor of UTI (P = 0.023, OR = 5.93; 95% CI: 1.28–27.52). In fact, the stricture is a well-known risk factor for UTI. The reported rate of strictures after ureteroenteric anastomotic is about 10% (20), and most of which is probably caused by periureteral fibrosis or scarring secondary to ischemia after surgery (5, 21, 22). The formation of stricture contributed to infection, hydronephrosis, and even kidney failure (23). Therefore, for these patients, more closely follow-up and timely intervention are needed in order to preserve the renal function.

Controversy remains between the urinary diversion type and UTI risk. Parker et al. (10) proposed that continent urinary diversion resulted in significantly increased UTI risk. Likewise, a population-based study (15) in Sweden showed that patients who underwent ONB diversion have an increased risk of UTI compared with those who received IC reconstruction (OR = 1.21; 95% CI: 1.05–1.39). However, Clifford et al. (13) concluded that no significant difference in the 90-day UTI rate between the three diversion types (ONB, IC, and CCD). In this study, we noted that compared with CCD, IC, and ONB reconstruction increased UTI risk (OR = 6.955; 95% CI: 1.69–28.68; and OR = 4.355; 95% CI: 1.35–14.02, respectively). A potential explanation for this association may be the prolonged duration of indwelling urinary catheters in patients after ONB reconstruction. Lo et al. (24) demonstrated that indwelling a urinary catheter for every additional day increased the risk of UTI by 3–7%. Although IC seems to contribute more to UTI than ONB, it cannot be underestimated that the cumulative incidence of UTI in patients with ONB reconstruction increased over the time.

Interestingly, PNI, a nutrition and inflammation-related index was identified as a predictor for UTI in our nomogram. Previous studies (25–27) have evaluated the prognostic value of PNI in many cancers. To date, no studies have explored the predictive ability of PNI in UTI. In this work, we demonstrated that a comprehensive model comprised of PNI enhanced the predictive ability of UTI, as shown by improvement in NRI and IDI. Traditionally, BMI and preoperative serum levels of albumin were used to predict the risk of UTI. The calculation of PNI is based on preoperative albumin levels and our findings support prior research that preoperative nutrition status could be a predictor of postoperative UTI.

Urinary tract infection had a great impact on the length of hospital stay and increased morbidity and cost after RC (28). Several efforts have been made to prevent this common postoperative complication. Large et al. (29) tried to abrogate mechanical bowel preparation in patients who underwent RC and urinary diversion, however, the results showed no significant difference between the two groups in UTI rate (p = 0.6). Kolwijck et al. (30) suggested that the antibiotic prophylaxis should be broadened for patients undergoing RC based on relatively high incidence of infection (18.4%), especially when the ureteral stent was removed. Wang et al. (31) found that based on the results of postoperative urine culture of stub of J-stent culture in 3 and 7 days after surgery, antibiotic prophylaxis usage significantly decreased the UTI rate (p = 0.031).

There are several limitations of this study. First, this study was confined by its retrospective nature. Second, the nomogram was established based on data obtained from a single institution in China and was validated internally, restricting the promotion of our developed model. Third, the results of antibiotics susceptibility experiments exist in vitro, we are unable to detect the efficacy and side effect of selected drugs. Therefore, more well-designed prospective studies and multicenter studies are warranted.

In conclusion, this study introduces an early warning model to predict the incidence of 90-day UTI after RC and urinary diversion in patients with bladder cancer. The nomogram shows great predictive accuracy.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

This study was approved by the Local Institutional Review Board and was censored on June 30, 2021. Informed consent was waived due to the retrospective nature.

XL, HJ, and DW conceived, designed the experiments and contributed to the interpretation of the data. XL, YW, QC, and SC participated in the experiments and drafted the manuscript. QC and XL performed the statistical analysis. SC and MC revised the manuscript. All authors read and approved the final version of the manuscript.

This study was supported by the National Natural Science Foundation of China (Nos. SQ2017YFSF090096 and 82070773) and the Natural Science Foundation of Jiangsu Province (BE20201271).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.