The threshold effect of factors associated with repeated implantation failure in human‑assisted reproductive technology

Abstract

Background In assisted reproductive technology (ART), repeated implantation failure is bound to have a higher potential risk when compared to first successful clinical pregnancy in ART. However, few studies have comprehensively analyzed the influence of pre-ART cycle related factors on repeat implantation failure. Therefore, our aim was to identify the influencing factors for RIF and explore different new risk factors and cutoff values before ART. Methods This was a retrospective cohort study of 1581 patients successfully collected with ART in Qinghai Province people’s Hospital and Lanzhou University Second Hospital from January 2016 to December 2020. Overall, 671 (76%) first successful clinical pregnancy group in ART had an and 211 (24%) repeated implantation failure. We explored the independent risk factors associated with cases of RIF in infertile patients treated with human‑assisted reproductive technology (ART) and established a smooth curve fit and perform a threshold effect analysis can provide guidance and a valuable reference for predicting the probability of RIF. Results The overall RIF rate in current ART population was 24%. Among the 882 ART cycles, a total of 671 ended as first successful clinical pregnancy, 211 as repeated implantation failure. The Age of female (32(29,36) vs. 30(28,36)), P < 0.001)) was significantly different between RIF group and first clinical pregnancy group. In the ART cycle, WBC (5.90(4.73, 7.40) vs. 5.70 (4.70, 6.78) P = 0.044), NEUT (3.66(2.80, 5.10) vs. 3.44(2.70, 4.32), P = 0.008), MCH (30.80 (29.70, 31.80) vs. 30.40 (29.20, 31.50), P = 0.008), MCHC (336 (328, 342) vs. 333 (327, 340), P = 0.013) ,and RDW-SD (43.3 (42, 45.1) vs. 43 (41.6, 44.6), P = 0.006) in RIF group were higher than in first clinical pregnancy group, and there were statistical differences between the two groups. In the ART cycle, INR (0.95(0.89,1.00) vs. 0.93 (0.88, 0.99) P = 0.049), APTT(27.70(24.8, 31.7) vs 26.40 (24, 29.60), P < 0.001), FIB (2.57 (2.22, 2.98) vs.2.50 (2.19, 2.86), P = 0.025) in RIF group were higher than in first clinical pregnancy group, and there were statistical differences between the two groups. In ART cycle, TT (18.40 (14.40, 19.40) vs.18.80 (17.7, 19.7), P < 0.001), DBIL (3.40(2.20, 4.40) vs. 3.70 (2.70, 4.80) P = 0.002), and FT3 (5.03(4.67, 5.50) vs. 5.22 (4.79, 5.64) P = 0.002), in RIF group were lower than in first clinical pregnancy group, and there were statistical differences between the two groups. Univariate regression analysis showed that female age, WBC, NEUT, MCHC, RDW-SD,DBIL,APTT,FIB,TT has statistical difference between RIF group and first clinical pregnancy group. Multivariate logistic regression analysis showed that female age(odds ratio [OR] = 1.105, 95%CI = 1.064–1.148, P < 0.001), MCHC(odds ratio [OR] = 1.028, 95%CI = 1.013–1.044, P < 0.001), RDW-SD(odds ratio [OR] = 1.096, 95% CI = 1.037–1.158, P = 0.001), FIB (odds ratio [OR] = 0.827, 95%CI = 0.757–0.904, P < 0.001) has statistical difference between RIF group and first pregnancy group. The threshold effect analysis found that the threshold point for female age was 31.5 years (cut-off point),when female age ≥ 31.5 years old, the risk of RIF increases sharply with age (OR2.221; 95% CI = 1.623–3.040; P < 0.001). The correlation between female age and the probability of RIF tends to be more stable; the older the age, the greater the probability of RIF. The threshold point of MCHC is 338.5 g/L (cut-off point) ,when MCHC ≥ 338.5 g/L the risk of RIF increases sharply with MCHC(OR 1.727; 95%CI = 1.256–2.375; P = 0.001).The correlation between MCHC and the probability of RIF tends to be more stable; the higher the MCHC, the greater the probability of RIF.The threshold point of RDW-SD is 42.7 fl (cut-off point) ,when RDW-SD ≥ 42.7 fl ,the risk of RIF increases sharply with RDW-SD (OR 1.547; 95% CI = 1.121–2.135; P = 0.008).The correlation between RDW-SD and the probability of RIF tends to be more stable; the higher the RDW-SD, the greater the probability of RIF. The threshold point of FIB is 2.759 g/L (cut-off point), when FIB ≥ 2.759 g/L the risk of RIF increases sharply with FIB, the risk of RIF increases sharply with FIB (OR 1.531; 95% CI = 1.112–2.108; P = 0.009). The correlation between FIB and the probability of RIF tends to be more stable; the higher the FIB, the greater the probability of RIF. Conclusion The age of both women, MCHC, RDW-SD and FIB were statistically different between the patients with first clinical pregnancy group and the patients with repeated implantation failure group. As women age, the probability of ART clinical pregnancy decreases. The lower the MCHC, the lower the RDW-SD and the lower the FIB, the higher the probability of clinical pregnancy in ART patients. We found that MCHC, RDW-SD and FIB affect the occurrence of RIF in infertile women. Our findings suggest a preliminary therapeutic mechanism for the clinical anticoagulation therapy of RIF.