OBJECTIVE: To systematically and meta-analytically pool the existing evidence regarding the prognostic impact of preoperative anemia (hemoglobin level <12 mg/dl) in patients with endometrial cancer.
METHODS: Four (PubMed, Embase, Scopus and Web of Science) databases were searched from inception to 20-August-2020. We assessed the risk of bias using the Newcastle-Ottawa Scale. We estimated the pooled prevalence of preoperative anemia in the included studies. We pooled odds ratios (ORs) and hazard ratios (HRs) with their 95 % confidence intervals (95 % CIs) to evaluate the correlation between preoperative anemia and its impact on
clinicopathologic parameters and survival outcomes. Analyses were performed under random- or fixed-effects meta-analysis models depending on data heterogeneity.
RESULTS: Seven studies met the inclusion criteria comprising 1495 patients with endometrial cancer. Nearly all studies had low risk of bias. The pooled prevalence of preoperative anemia was 26.5 % (95 % CI: 18.6%-36.2%). Preoperative anemia significantly correlated with advanced FIGO stage III-IV (OR = 5.14, 95 % CI [3.36, 7.86], p < 0.00001), ≥50 % myometrial invasion (OR = 1.95, 95 % CI [1.36, 2.78], p = 0.0003), lymph node metastasis (OR = 4.46, 95 % CI [2.39, 8.30], p < 0.00001), non-endometrioid histology (OR = 3.25, 95 % CI [1.89, 5.60], p < 0.0001), adnexal involvement (OR = 5.88, 95 % CI [3.05, 10.23], p < 0.001), cervical involvement (OR = 2.91, 95 % CI [1.65, 5.11], p = 0.0002), positive peritoneal cytology (OR = 3.24, 95 % CI [1.41, 7.44], p = 0.006), preoperative thrombocytosis (OR = 6.66, 95 % CI [3.05, 14.52], p < 0.00001) and lymphovascular space invasion (OR = 3.50, 95 % CI [1.82, 6.74], p = 0.0002). High tumor grade II-III was increased in patients with preoperative anemia, yet this effect was not statistically significant (OR = 2.12, 95 % CI [0.97, 4.66], p = 0.06). Consistently, the five-year overall survival (OS) and disease-free survival (DFS) rates were significantly lower in patients with preoperative anemia when compared to those without preoperative anemia. Pooled HR showed that preoperative anemia was significantly associated with reduced DFS at univariate (HR = 3.22, 95 % CI [1.28, 8.11], p = 0.01) and multivariate (HR = 1.02, 95 % CI [1.00, 1.05], p = 0.03) analyses.
CONCLUSIONS: Preoperative anemia predicts poor
clinicopathologic and survival outcomes in patients with endometrial cancer.