Secondary predictors of ovarian hyperstimulation syndrome development

Materials and methods. The study included 240 women at high risk of developing ovarian hyperstimulation syndrome, 63 of these women subsequently developed ovarian hyperstimulation syndrome, and 177 did not. To identify secondary predictors, a number of preovulatory follicles and oocytes was evaluated, estradiol levels, vascular endothelial growth factor (VEGF) and interleukin-6 (IL-6) were defined in the blood serum and follicular fluid.

Infertility is the most important medico-social problem not only in Ukraine, but throughout the world [1][2][3][4]. An effective method of infertility treatment is fertilization in vitro [1,5]. However, its widespread introduction into clinical practice does not only help to solve the problems of infertility, but also can be accompanied by severe complications that threaten the health and even the life of the patient [5,6]. One of these complications is ovarian hyperstimulation syndrome (OHSS) [5][6][7][8][9][10].
Despite the large number of studies on the given state, up to the present date, there is no perfect strategy that completely eliminates OHSS. At the same time, there are risk factors that can be taken into account in order to reduce the frequency of its development [7][8][9][10]. Primary predictors of OHSS development, including the number of antral follicles, antimullerian hormone levels, polycystic ovary syndrome, OHSS in previous cycles, young age and low body weight in patients are proved by many scientists and already widely used in practice while selecting controlled ovarian stimulation [7][8][9][10]. However, despite the implementation of preventive measures, the risk of developing OHSS remains because of the controlled ovarian stimulation. Therefore, identification and evaluation of OHSS secondary predictors allow to develop further tactics of patient treatment that will contribute to the reduction of OHSS cases up to complete elimination.

Aim
To identify and evaluate secondary predictors of OHSS in women at high risk of OHSS.

Materials and methods
The study included 240 women at high risk of OHSS development taking into account the primary predictors such as: the number of antral follicles more than 20, the content of antimullerian hormone 5.5 ng/ml and more, polycystic ovary syndrome, OHSS during previous in vitro fertilization attempts. Subsequently, 63 of these women developed OHSS and 177 did not.
An echographic study was performed using an ultrasound diagnostic system SSA-790A (APLIO-XG) (Toshiba, Japan) with transabdominal and transvaginal convex transducers.
Definition of estradiol, vascular endothelial growth factor (VEGF) and interleukin-6 (IL-6) in the blood serum and follicular fluid were performed on the day of transvaginal follicle puncture. The follicular fluid was obtained from the dominant follicles, centrifuged for 5 minutes and separated from the blood cells. Using the electrochemiluminescent method and the reagents of Roche Diagnostics (Switzerland) company on an automated analyzer Cobas-e411 (Roche Diagnostics, Switzerland), estradiol concentration was determined; by immune-enzyme method using an immune-enzyme reader ImmunoChem-2100 (Hight Tehnology, USA), VEGF was investigated using the reagents of the company "BIO Tech Lab S" (USA) and IL-6 -of "Vector-Best" company (RF).
Statistical processing and the obtained data analysis were performed using the SPSS Statistics 22 software package. Since the most of the variables were not normally distributed, they were represented by the median (Me) and interquartile ranges -25 and 75 percentiles (25; 75 %). The comparison of indicators between groups was performed using the non-parametric Mann-Whitney U test. To determine the direction and strength of a relationship between the variables the Spearman rank coefficient of correlation (r s ) and the 95 % confidence interval (95 % CI) was further calculated.
Binary logistic regression method was used to determine the boundary values of independent variables relative to the probability of OHSS occurrence. The prognostic test sensitivity (true positive result) and the specificity of the test (true negative result) were determined. The ROC-curves were represented. The value, which was diagnosed with a zero degree of prognosis, was depicted in the figures as a diagonal dashed line. Consequently, the more steeply curved ROC-curve (the maximum degree of prediction 1) was, the more accurate was the test results prediction.
Correlation dependence between the follicular fluid and blood serum variables indicates a tendency to increase or decrease simultaneously, which, according to our opinion, in the case of unavailability of all laboratory values determination, makes it possible to determine only the indicators available in a particular laboratory, serum or follicular fluid.
By the method of binary logistic regression it was determined that the boundary value in OHSS was the preovulatory follicles number ≥30 for two ovarian, the sensitivity of the given predictor was 46.0 % and the specificity reached 91.0 %. The number of oocytes obtained after transvaginal puncture is one of the important predictors for OHSS occurrence. It was established that OHSS development could be assumed with the number of oocytes ≥19 with a sensitivity of 87.3 % and a specificity of 97.7 %.
In the course of a further study it was found that the boundary value for OHSS development was estradiol content in the blood serum ≥4800 pg/ml and in follicular fluid ≥725 000 pg/ml. The sensitivity for these predictors was 74.6 % and 98.4 %, the specificity was 96.6 % and 97.7 %, respectively. The next indicator analyzed was VEGF, the boundary value of which for OHSS development corresponded to ≥30 pg/ml in the blood serum and ≥500 pg/ml in the follicular fluid. The sensitivity for these predictors was 76.2 % and 87.3 %, the specificity was 98.3 % and 97.2 %, respectively.
By the further analysis determined that the patients with IL-6 level in the blood serum ≥4 pg/ml and ≥8.5 pg/ml in the follicular fluid were classified as OHSS category. The sensitivity was 65.1 % and 90.5 %, the specificity -98.9 % and 98.3 %, respectively.
Developing ROC-curves pointed to the high diagnostic efficiency of the considered predictors (Fig. 4). Thus, the area under the ROC-curve for preovulatory follicles

Discussion
Thus, on the basis of the conducted study, secondary predictors of OHSS development include: the number of preovulatory follicles ≥30 per two ovaries, the number of oocytes ≥19, the level of estradiol in the blood serum ≥4800 pg/ml and in the follicular fluid -≥725000 pg/ml, VEGF -≥30 pg/ml and ≥500 pg/ml, IL-6 -≥4 pg/ml and ≥8,5 pg/ml, respectively.
In identifying secondary predictors of OHSS development, the main management tactics for such patients is discontinuation of luteal phase support, vitrification of all embryos and their transfer in subsequent thawed cycles.