Clinical and prognostic value of preoperative hydronephrosis in upper tract urothelial carcinoma: a systematic review and meta-analysis

Literature research

This meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement (Moher et al., 2010). A systematic literature search using PubMed, ISI Web of Knowledge, and Embase was conducted to retrieve clinical studies up to March 1, 2016. The search terms used included the following: “ureteral neoplasms,” “urothelium,” “ureter,” “renal pelvis,” “upper tract urothelial,” “hydronephrosis,” “preoperative hydronephrosis,” “prognosis or prognostic or outcome,” and relevant variants of these search terms. The following criteria were used to determine study eligibility: (1) must concern the connection between preoperative hydronephrosis and clinicopathological characteristics and the prognosis of UTUC; (2) UTUC patients must have been diagnosed using the standard histopathological examination criteria; and (3) must provide information about the preoperative hydronephrosis. The exclusion criteria for the study were: (a) studies lacking original data; (b) not written in English; and (c) reviews, meta-analyses, case reports, abstracts, or meeting records. For overlapping articles, we included the most informative and latest article.

Data extraction and quality assessment

All data were independently reviewed by two researchers (YWG and YYP) and were cross-checked. Additionally, any disagreement or uncertainty was resolved using group discussion. The quality of the selected articles was assessed according to the Newcastle-Ottawa scale (NOS) criteria (Stang, 2010). For quality, scores ranged from 0 (lowest) to 9 (greatest); studies with scores of 5 or more were graded as good quality. The data extracted from these citations included the name of the first author, publication year, country, number of patients, recruitment period, sex ratio, age, cut-off, prognostic outcomes, pathological stage, lymph node dissection, tumor grade, and tumor location. The data were extracted from the original articles. Situations lacking exact data were resolved in a number of ways: multivariate outcomes were used before univariate outcomes when both were presented, but if no multivariate results were presented, univariate outcomes were used instead.

Statistical analysis

Odds ratios (ORs) and 95% CIs were used to estimate the relationships between preoperative hydronephrosis and clinicopathological parameters, including pathologic tumor stage, tumor grade, lymph node status, tumor location, recurrence status, and gender. HRs and 95% CIs were used to evaluate the relationships between preoperative hydronephrosis and cancer-specific survival (CSS), overall survival (OS), recurrence-free survival (RFS), or metastasis-free survival (MFS). A p-value of < 0.05 was considered statistically significant. The statistical significance of the pooled ORs and HRs was evaluated using a Z-test. Heterogeneity among studies was evaluated using the Cochran Q-statistic and the I² test (Zintzaras & Ioannidis, 2005). A random effects model was used when significant heterogeneity existed among studies (P < 0.05 or I² > 50%); otherwise, a fixed-effects model was used. Funnel plots and Begg’s test were used to evaluate the potential publication bias (Peters et al., 2006). All statistical calculations were performed using Review Manager 5.3 (The Cochrane Collaboration, Copenhagen) and STATA version 14.0 (Stata Corp, College Station, TX).

Eligible studies and quality assessment

Initially, our search strategy identified 357 articles. A total of 19 studies published between 2007 and 2016 were included in the final meta-analysis (Bozzini et al., 2013; Chapman et al., 2009; Chen et al., 2013; Cho et al., 2007; Chung et al., 2014; Colin et al., 2014; Fradet et al., 2014; Hwang et al., 2013; Liang et al., 2016; Luo et al., 2013; Messer et al., 2013; Ng et al., 2011; Sakano et al., 2013; Xing et al., 2016; Yeh et al., 2015; Zhang et al., 2016; Zhang et al., 2013; Zhang et al., 2015; Zou et al., 2014) (Fig. 1). Thirteen studies provided original information of the relationships between preoperative hydronephrosis and the clinicopathological parameters of UTUC patients directly (Bozzini et al., 2013; Chapman et al., 2009; Chen et al., 2013; Cho et al., 2007; Chung et al., 2014; Fradet et al., 2014; Hwang et al., 2013; Liang et al., 2016; Messer et al., 2013; Ng et al., 2011; Yeh et al., 2015; Zhang et al., 2013; Zhang et al., 2015). Fourteen articles assessed the prognostic value of preoperative hydronephrosis (CSS/OS/RFS/MFS) of UTUC patients using the Kaplan–Meier method (Bozzini et al., 2013; Chapman et al., 2009; Chung et al., 2014; Colin et al., 2014; Hwang et al., 2013; Liang et al., 2016; Luo et al., 2013; Ng et al., 2011; Sakano et al., 2013; Xing et al., 2016; Yeh et al., 2015; Zhang et al., 2016; Zhang et al., 2013; Zhang et al., 2015). The main characteristics of the 19 studies included in our meta-analysis are shown in Table 1. Other characteristics, such as the pathological results and the prognostic features, are summarized in Tables S1, S2, and S3. A total of 5,782 patients were included in this meta-analysis, and they were from 7 countries (Korea, USA, Japan, France, China, Canada, Taiwan). The median follow-up period of the studies ranged from 1 month to 233 months. The age of the patients ranged from 20 to 95 years, and the overall proportion of males was 51.74%.

Present/high preoperative hydronephrosis was defined using radiographic reports. Differences in the cut-off value of present/high preoperative hydronephrosis were observed among the studies. Preoperative hydronephrosis that was present or a high grade was considered to be positive, and absent or a low grade was considered to be negative.

Fourteen articles evaluated the prognostic value of preoperative hydronephrosis (CSS/OS/RFS/MFS) in UTUC patients. Of the 14 studies, 12 provided HR and 95% CI values directly; of the other two studies, one paper provided the relative risk (RR), and the other article provided OR values, which we used to estimate HR. Of the 14 studies, a significant association between preoperative hydronephrosis and poor CSS, OS, RFS or MFS was demonstrated in six (Chung et al., 2014; Liang et al., 2016; Ng et al., 2011; Yeh et al., 2015; Zhang et al., 2013; Zhang et al. 2015), four (Chapman et al., 2009; Liang et al., 2016; Yeh et al., 2015; Zhang et al., 2015), three (Chung et al., 2014; Hwang et al., 2013; Luo et al., 2013) or one (Ng et al., 2011) studies, respectively. Of the literature, the six (Bozzini et al., 2013; Sakano et al., 2013; Xing et al., 2016; Yeh et al., 2015; Zhang et al., 2016; Zou et al., 2014), two (Bozzini et al., 2013; Yeh et al., 2015), four (Chung et al., 2014; Liang et al., 2016; Luo et al., 2013; Ng et al., 2011) or three (Bozzini et al., 2013; Colin et al., 2014; Luo et al., 2013) studies linking preoperative hydronephrosis with poor CSS, OS, RFS or MFS, respectively, lacked statistical significance.

Survival outcomes

Of the 12 studies investigating the association between preoperative hydronephrosis and CSS, the pooled HR and 95% CI for UTUC patients was 1.69 (95% CI [1.23–2.33], P = 0.001, n = 3,063) with heterogeneity (I² = 70%, P = 0.0001; Table 2 and Fig. 2A). The pooled HR and 95% CI for OS provided in six studies was 1.62 (95% CI [1.35–1.94], P < 0.00001) with heterogeneity (I² = 17%, P = 0.30; Table 2 and Fig. 2B).

The results also demonstrated significant associations between presence of preoperative hydronephrosis and shorter RFS and MFS, respectively; the combined HRs were 1.95, 95% CI [1.26–3.04], P = 0.003 and 1.55, 95% CI [1.04–2.33], P = 0.03, respectively (Table 2, Figs. S3A and S3B).

Relationships between preoperative hydronephrosis and clinicopathological parameters

In this meta-analysis, clinicopathological features such as tumor stage, tumor grade, lymph node status, tumor location, recurrence status, and gender, as impacted by the presence of preoperative hydronephrosis, were compared using the 13 studies. The results of the meta-analysis showed significant associations between presence of preoperative hydronephrosis and higher tumor stage (T2-4), positive lymph node metastatic status or ureter tumor location, respectively; the combined ORs and 95% CIs were as follows: OR 3.12, 95% CI [1.85–5.27], P < 0.0001; OR 1.60, 95% CI [1.13–2.26], P = 0.009; OR 3.75, 95% CI [2.44–5.77], P < 0.00001 (Table 3, Figs. S1A, S2A and S2B). Moreover, significant associations between preoperative hydronephrosis and tumor grade (high grade vs. low grade) were observed only in UTUC patients. The OR and 95% CI was as follows: OR 1.66, 95% CI [1.20–2.29], P = 0.002 (Table 3 and Fig. S1B). There was no significant association between preoperative hydronephrosis and tumor grade (G3 vs. G1 + G2 and G2 + G3 vs. G1, respectively); the combined ORs and 95% CIs were OR 1.42, 95% CI [0.75–2.69], P = 0.28 and OR 0.94, 95% CI [0.49–1.79], P = 0.85 (Table 3, Figs. S1C and S1D]).

Finally, there was no significant association between preoperative hydronephrosis and recurrence status (bladder recurrence vs. no bladder recurrence) or gender (male vs. female), respectively; the combined ORs and 95% CIs were OR 1.28, 95% CI [0.92–1.77], P = 0.14 and OR 1.01, 95% CI [0.83–1.22], P = 0.927 (Table 3, Figs. S2C, and S2D).

Publication bias

Publication bias was assessed using Begg’s test and Egger’s test for asymmetry only for cancer-specific survival (CSS) of UTUC (Figs. 3A and 3B). No evidence of asymmetry was found using our funnel plot. Begg’s test (P = 0.244) and Egger’s test (P = 0.093) suggested that our analyses were stable.