Plasma pentraxin 3 is higher in early ovarian hyperstimulation syndrome than in uncomplicated in vitro fertilization cycle of high-risk women

Abstract

Purpose: Pentraxin 3 (PTX3) is a locally secreted, quicker responsive pro-inflammatory protein than C-reactive protein (CRP). We evaluated the value of PTX3 in the prediction of ovarian hyperstimulation syndrome (OHSS), a severe complication of in vitro fertilization (IVF).

Methods: This two-year prospective follow-up study included 27 women with uncomplicated IVF-cycles (IVF group) and 31 patients diagnosed with moderate or severe early OHSS (OHSS group). PTX3 was analysed from follicular fluid (FF) and serial blood samples with enzyme-linked immunoassay and CRP with particle-enhanced immunoturbidimetric assay. The value of PTX3 and CRP in detecting OHSS was examined with receiver operating characteristic (ROC) curve analysis and expressed as the area under the curve (AUC).

Results: The circulating PTX3 level peaked at two days after oocyte pick-up (OPU2), and in the OHSS group the level was 1.9 times higher (P = 0.006) than in the IVF group. However, in ROC curve analysis PTX3 (AUC 0.79, best cut off 1.1 µg/L) was not superior to CRP (AUC 0.87; best cut off 9.5 mg/L) in predicting early OHSS. In the IVF group, the FF-PTX3 concentration was 15-20 times higher than in the plasma. PTX3 level at OPU2 correlated with the number of punctured follicles (r = 0.56, n = 22, P = 0.006). Triggering with human chorionic gonadotrophin or early pregnancy had no effect on PTX3 level.

Conclusion: The elevated PTX3 concentration in OHSS at OPU2, when freeze-all embryos strategy is still possible to consider, indicates that PTX3 level could provide additional benefit in the risk assessment for early OHSS.

Fig. 1

a Plasma pentraxin 3 b C-reactive Protein and c anti-Müllerian hormone concentrations during uncomplicated in vitro fertilization (IVF) cycle and early ovarian hyperstimulation syndrome (OHSS). The data are expressed as medians and 95% confidence intervals. Pairwise comparisons were done with Wilcoxon Signed Rank test (Friedman’s presumption was significant) with Bonferroni correction: *P < 0.05, **P ≤ 0.01 (compared with baseline, when not otherwise stated); and a**P ≤ 0.01, Mann–Whitney U test between IVF cycle and OHSS. OPUd, days after the oocyte pick-up

Fig. 2

Receiver operating curves of plasma pentraxin 3 and C-reactive protein for detecting early ovarian hyperstimulation syndrome (OHSS), two to three days after oocyte pick-up. Sensitivity and specificity for detecting early OHSS were a* 72% and 79% for PTX3 and b* 75% and 77% for CRP respectively. AUC, area under the curve

Fig. 3

Plasma pentraxin 3 (PTX3) and C-reactive protein (CRP) levels during the recovery from early ovarian hyperstimulation syndrome (OHSS). The time points on X axis: A on admission to the emergency department; W the day of worst symptoms; D at discharge from the hospital; S at the surveillance visit one week after the discharge. The data are expressed as medians and 95% confidence intervals. Pairwise comparisons were done with Wilcoxon Signed Rank test (Friedman’s presumption was significant): *P < 0.05; **P ≤ 0.01, compared with the value on admission, when not otherwise stated

Fig. 4

Follicle fluid and plasma pentraxin 3 and anti-Müllerian hormone levels at oocyte pick-up in uncomplicated IVF cycle and early ovarian hyperstimulation syndrome (OHSS). The data are expressed as medians (line inside the box), inter-quartile range (the box), the first and fourth quartile (whiskers). Logarithmic scale on Y-axis. Open dots represent outliers. Comparisons between groups are done with Mann–Whitney U test or Wilcoxon signed Rank test: *P < 0.05; **P ≤ 0.001