Impact of CRP on continuous bladder irrigation in hemorrhagic cystitis patients after hematopoietic stem cell transplantation

Background Continuous bladder irrigation (CBI) and proper adjustment of saline irrigation speed are important to avoid CBI failure in hemorrhagic cystitis (HC) patients after allogeneic hematopoietic stem cell transplantation (HSCT). Nevertheless, too fast irrigation speed could take away the patient's much heat, contribute to blood coagulopathy, and increase the nursing workload. Evaluation of risk for CBI failure remains an unmet clinical need. Methods The general information, clinical characteristics, and consultation records of HC patients in 1380 patients with hematopoietic stem cell transplantation in our center from 2017 to 2019 were analyzed retrospectively. The receiver operating characteristic (ROC) curve was used to calculate the cutoff point of the continuous variable, and multivariate logistic regression was used to analyze the risk factors affecting CBI failure in HC patients. Results The incidence of HC after HSCT was 23%. A total of 227 patients with HC above grade 2 were included. Univariate analysis showed that CRP, age, platelet counts, onset time after transplantation, albumin, and hemoglobin were associated with CBI failure in the short-term (P < 0.05). ROC curve and multivariate logistic regression analysis showed that CRP > 8.89 ng/ml (RR = 7.828, 95% CI 2.885– 21.244), age < 14.5 years (RR = 9.940, 95% CI 3.219–30.697), and onset time of HC (cid:0) 37d after transplantation (RR = 7.021, 95% CI 2.204–22.364), were independent risk factors for failure of CBI (P < 0.05). Conclusions The study identied CRP > 8.89 ng/ml, age < 14.5 years, and onset time of HC after HSCT > 37d are independent factors for failure of CBI, which could be combined to allow stratication of HC after HSCT patients into low-, intermediate- and high-risk subgroups of CBI failure.


Background
Hemorrhagic cystitis (HC) after allogeneic hematopoietic stem cell transplantation (HSCT) is characterized by diffuse in ammation and hemorrhage of the bladder mucosa. Its clinical manifestation, severity, and prognosis vary greatly. It has been reported that the incidence of HC, as one of the major complications in allogeneic HSCT, is 14% − 30% [1][2][3][4]. Refer to the Droller , s HC classi cation (Table 1), grade I means only microscopic haematuria, and gross hematuria means grade or higher. Conservative observation, hydration, alkalization of urine, diuretics, and antiviral therapy were e cient for most HC patients with grade I or II, while continuous bladder irrigation (CBI) was required for some grade , III, and IV patients to avoid urinary tract obstruction caused by blood clots in the bladder. Gross hematuria with massive clotting, causing urinary tract obstruction, requiring instrumentation for clot evacuation Patients with this allogeneic HSCT have abnormal immunity, coagulopathy [5], and graft-versus-host disease (GVHD). For urinary tract obstruction of HSCT patients, surgical treatment is associated with mortality and effects were minimal. CBI and proper adjustment of saline irrigation speed are important to avoid CBI failure. Nevertheless, too fast irrigation speed could take away the patient's much heat, contribute to blood coagulopathy, and increase the nursing workload.
Doctors often believe that the grading of HC is the most reliable factor to predict the failure of CBI for HC, which is not practical in clinical practice. Many patients were diagnosed with grade IV of HC as bladder irrigation failed. It means HC grade classi cation of the moment is not of enough practical signi cance in guiding the saline irrigation speed for HC to prevent CBI failure. Therefore, it is an unsolved clinical need to predict reliably the CBI failure of HC. We hypothesized that patients with HC grade II or higher had some clinical characteristics to predict the risk of CBI failure. We report an analysis of HC incidence and treatment outcome in our center.

Materials And Methods
The study was approved by the institutional review board in our center. The general information, clinical characteristics, and consultation records of HC patients in 1380 patients with HSCT in our center from 2017 to 2019 were analyzed retrospectively. According to the consultation records in our center, if the irrigation line was completely blocked in CBI, it is determined as CBI failure of the patient. The following clinical parameters were collected: general information, the primary diseases, onset time of HC after HSCT, haploidentical HSCT, sex-mismatch in recipients, cytomegalovirus (CMV) viremia, EB viremia, hemoglobin, platelets, serum creatinine, C-reactive protein (CRP), and serum albumin in the presence of gross hematuria.
SPSS 26.0 statistical software was used for data analysis and data processing. The receiver operating characteristic (ROC) curve was used in measurement data to determine the appropriate cut-off point of predicting CBI failure, by which measurement variables were divided into binary variables. Univariate analysis and multivariate logistic regression were used to analyze the risk factors affecting CBI failure in HC patients. The difference was statistically signi cant at 0.05.

Establishment of the cut-off value of the continuous variable
To determine the cutoff points of CRP, age, platelets count, post-transplantation onset time, serum albumin, and hemoglobin in the entire cohort, the ROC curve for CBI failure was plotted (Fig. 1)

Discussion
Urologists are often bothered by hematologists' consultation about the failure of CBI in HC patients, which may require manual processing of blood clots and even surgical intervention in patients with immunity de ciency and hematopoiesis. In grade II, III, and all IV of HC patients related to HSCT, CBI is suggested to prevent large blood clotting in bladder forming and lower urinary tract obstruction to allow adequate catheter drainage. Although adjusting the speed of bladder saline irrigation can avoid the failure of CBI, medical nursing providers always slow down the speed of irrigation in any grades of HC patients for various reasons limiting CBI prophylactic value, such as hypothermia [4,6]. Notably, if CBI failure was not resolved in time and CBI was not suspended, iatrogenic rupture of the bladder or renal function damage could be caused. Evaluation of risk for CBI failure remains an unmet clinical need. Compared with radiation cystitis, although the pathological changes and symptoms of HC related to HSCT are mainly in the bladder, its diagnosis and treatment can not bypass the special immune de ciency and abnormal hematopoietic function of HSCT, which means higher signi cant morbidity and mortality in surgical treatment. Through a large sample retrospective study in our center, the data of 1380 patients who received HSCT were analyzed retrospectively. The incidence of HC after HSCT was 23% in our center. The CRP, platelet count, and hemoglobin of patients with gross hematuria were analyzed statistically, and the independent risk factors of CBI failure were discussed. In the study, CRP was rstly found to be an independent risk factor for CBI failure, and its cutoff point CRP 8.89 mg/dl was determined. Up to the date, CRP has not been reported as independent risk factors for HC grade or treatment outcome. In the current study, we have con rmed that patients with the onset time > 37d of HC after transplantation are more likely to fail in CBI. This is consistent with previous studies. Johnston et al found that delayed HC onset time was an independent risk factor for CBI of HC in children [1]. Our study further determined the cutoff point and con rmed and that HC onset time 37d is still an independent risk factor for CBI failure.
Within 3 days after the pretreatment of allogeneic HSCT, HC was mainly contributed to the toxicity of cyclophosphamide (CY) in the pretreatment. When acrolein, the metabolite of CY, combines with the epithelial tissue of bladder mucosa, it will cause extensive damage to the mucosa of the patients. At the same time, acrolein is more likely to form crystals and deposits in the kidney or bladder in the acid environment, which aggravates the bleeding and necrosis of the mucosa. Nevertheless, previous studies showed that the HC after 3 days of HSCT was mainly related to the infection of CMV, adenovirus, BKV [4], JC virus [7], and so on. Stanchi [8] et al. reviewed adverse events of 214 recipients undergoing HSCT in children, and the results indicated that CRP increased signi cantly during sepsis / SIRS, bacteremia graft rejection, liver or intestinal GVHD, and viremia. In this study, CRP in hematuria was found to be an independent risk factor for CBI failure in HC after HSCT, and CRP 8.89 mg/dl of gross hematuria stage is superior to single virus detection.
The urethra is more narrow in younger patients, which is closely associated with the CBI failure of HC. Lucila et al [9]. reviewed 133 recipients undergoing HSCT and found that younger patients were more likely to develop BK polyomavirus related HC in multivariate regression analysis and had a worse prognosis. Nevertheless, in Johnston , s study, age as a continuous variable was not associated with HC treatment outcome(p = 0.0773) and HC grade(p = 0.0721) [1]. In this study, multiple regression analysis was carried out in the multi-age group of HSCT. It was found that the age of fewer than 14.5 years old was an independent risk factor for the failure of CBI in HC patients.
For the applicability of the research results, the study excluded grade I HC patients, for these patients had no gross hematuria or CBI. The limitations of this study are as follows: 1. This study is retrospective and all data was collected from a single-center. However, our center has received a large number of HSCT patients from all over the country. The source of HSCT patients is rather su cient. Clinically factors can still be collected from this single-center, which may guide the current treatment and future investigation.

Conclusion
HC, as one of the complications of HSCT, is not rare. Compared with radiation cystitis, HSCT related HC has unique clinical characteristics. The study identi ed CRP > 8.89 ng/ml, age < 14.5 years, and onset time of HC after HSCT > 37d are independent factors for failure of CBI, which could be combined to allow strati cation of HC after HSCT patients into low-, intermediate-and high-risk subgroups of CBI failure. In the clinical pathway, we can avoid the dilemma of surgical intervention by taking more targeted and active interventions. In the future, we need to expand the sample size and conduct the prospective, multicenter study to gradually improve the predicted model.

Consent for publication: Not applicable
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.