Stephen P. Boyers

Reduction of postoperative pelvic adhesions in the rabbit with Gore-Tex surgical membrane.

Gore-Tex (W.L. Gore and Associates, Inc., Flagstaff, AZ) surgical membrane (SM), a nonreactive expanded polytetrafluoroethylene (PTFE), was used in 24 mature New Zealand rabbits (2200 to 3000 gm) to cover 2-cm2 ischemic defects in the pelvic sidewall peritoneum to reduce adhesion formation in a rabbit pelvic sidewall/uterine horn injury model. SM was randomly assigned to cover one defect; the opposite defect remained uncovered, each animal serving as its own control. Rabbits were sacrificed 3 weeks later, and adhesions between uterine horn and pelvic sidewall or SM were scored for extent (0 to 4), type (0 to 4), and tenacity (0 to 3). Injury sites were removed en bloc for histologic study. The mean (+/- standard deviation [SD]) adhesion score for SM-covered lesions (4.3 +/- 1.8) was significantly lower than for controls (9.1 +/- 2.5) (P less than 0.001; Wilcoxon Signed Rank test). By histology, none of 24 SM-covered lesions demonstrated adhesions to the membrane itself, whereas 19 of the 24 control lesions showed dense adhesions to the injury site (P less than 0.001; chi-square test). By both gross and microscopic assessment, SM was nonadherent to the underlying sidewall defect in 100% of cases. In conclusion, Gore-Tex surgical membrane is an effective barrier for reducing primary adhesions in this pelvic injury model and offers promise for adhesion reduction in human pelvic surgery.

Human chorionic gonadotropin, estradiol, and progesterone profiles in conception and nonconception cycles in an in vitro fertilization program *

In 22 consecutive in vitro fertilization cycles stimulated with purified follicle-stimulating hormone, human chorionic gonadotropin (hCG), estradiol (E2), and progesterone (P) were measured every 3 days during the luteal phase. All serum measurements were normalized to the day of hCG administration (day 0). There was a total of nine pregnancies; two were biochemical pregnancies, whereas 7 of the 22 women had clinical pregnancies (31.8%). Of these, two miscarried and five had term pregnancies (three singleton, two twin). Conception cycles could be differentiated from nonconception cycles by serum E2 levels on day 8 (P = 0.035), by hCG levels on day 11 (P = 0.03), and by P levels on day 14 (P = 0.001). From days 8 to 11, hCG levels plateaued in conception cycles and decreased in nonconception cycles. However, during that period, E2 and P fell in both groups of women. This decline in sex steroids, which was observed in both conception and nonconception cycles, may well negatively influence endometrial development during the peri-implantation period and compromise conception, resulting in failure to conceive, biochemical pregnancy, and early miscarriage.

A paired analysis of in vitro fertilization and cleavage rates of first- versus last-recovered preovulatory human oocytes exposed to varying intervals of 100% CO2 pneumoperitoneum and general anesthesia.

This study compares the in vitro fertilization and cleavage rates of paired first- and last-recovered preovulatory human oocytes that were exposed to a 100% CO2 pneumoperitoneum and general anesthesia. In 305 consecutive cycles of laparoscopy, 1741 oocytes (5.7/cycle) were recovered. The exact time of aspiration (T) was recorded for each oocyte. The time interval (T1 to T2) between recovery of first and last oocytes ranged from 0 to 38 minutes and represented differences in the exposure time of first and last oocytes to the CO2 pneumoperitoneum and to general anesthesia. For all cycles (n = 305) without regard for T1 to T2, last-recovered oocytes fertilized less often than first-recovered eggs (P = 0.06; McNemars test). When T1 to T2 was short (less than or equal to 5 minutes), first- and last-recovered oocytes fertilized at comparable rates (70.8% and 74.0%). When only cycles with T1 to T2 greater than 5 minutes were considered (n = 209), the difference in fertilization rates between first and last oocytes (68.5% versus 56.4%) was highly significant (P less than 0.01; McNemars test). Pairing negated differences due to patient, cycle, or semen variables and first- and last-recovered oocytes had comparable maturity scores (4.0 +/- 0.5 versus 4.3 +/- 0.8). There were no significant differences in cleavage rates for first- and last-recovered oocytes that fertilized, regardless of the exposure interval (T1 to T2). We conclude that exposure to a 100% CO2 pneumoperitoneum and/or general anesthesia may adversely affect oocyte quality.

Isthmic ectopic pregnancy: segmental resection as the treatment of choice

We reviewed the records of 12 patients who were admitted to Yale-New Haven Hospital between February 1979 and January 1983 with the diagnosis of isthmic ectopic pregnancy. All pregnancies were unruptured. Two patients were managed by salpingectomy. Ten patients were treated conservatively. Of the women managed by conservative surgery, four had a linear salpingostomy and none of the four conceived. Three of these four patients demonstrated occlusion of the operated tube by hysterosalpingogram (HSG). The other six patients were managed by segmental resection and delayed microsurgical anastomosis. Four of the six patients conceived. Three pregnancies were intrauterine and one was an ectopic pregnancy in the conserved tube. Three patients conceived before an HSG could be done. The remaining three patients had HSGs 3 to 4 months after anastomosis, and the operated tube was patent in all three. From these data and a review of the literature, we conclude that segmental resection with either immediate or delayed anastomosis appears preferable to linear salpingostomy for the conservative management of unruptured isthmic ectopic pregnancy.

The effect of periovarian adhesions on follicular development in patients undergoing ovarian stimulation for in vitro fertilization-embryo transfer.

It has been suggested that the presence of periovarian adhesions might impair the ovarian response to gonadotropins. Periovarian adhesions were recorded in 49 women, and the total percentage of accessible ovarian cortex was described at the initiation of the operative procedure. Adhesiolysis was performed as needed for oocyte recovery. Ovarian access did not correlate with serum estradiol level on either the day of human chorionic gonadotropin (hCG) administration or the day after hCG administration. Similarly, neither the total number of follicles on the day of hCG or on the day after hCG, nor the number of follicles 1.0 to 1.4 cm or greater than or equal to 1.5 cm correlated with ovarian access. We conclude that periovarian adhesions are not a major determinant of the ovarian response to gonadotropin stimulation.

Hyaluronidase removal of the cumulus oophorus increases in vitro fertilization.

The effect of hyaluronidase removal of the cumulus oophorus on the in vitro fertilization rate of oocytes obtained from patients with poor oocyte fertilizability has been evaluated. Eighty-eight oocytes were obtained from 13 patients undergoing in vitro fertilization and embryo transfer (IVF-ET) for indications of male-factor, immunological, and idiopathic infertility. In addition, patients in whom fertilization did not occur on previous IVF cycles were evaluated in the study. The occytes of each individual patient were randomly assigned into a treatment (removal of the cumulus;N=40 oocytes) or nontreatment group (control;N=48 oocytes). Hyaluronidase was used to remove the cumulus immediately following oocyte retrieval, and insemination was performed 6–8 hr later. The overall oocyte fertilization rate (both treated and untreated) was 42%. The treatment group demonstrated a higher rate of fertilization compared to the nontreatment group (55% vs 31%;P<0.05). Examination of various patient groups revealed a statistically significant difference in fertilization rates between the treated and the untreated oocytes only in the “no previous fertilization” group (60% vs 28%;P<0.05). A higher rate of fertilization of the treated oocytes was also seen in the immunologic infertility group, however, statistical significance was not achieved (50% vs 25%;P=0.07). Only one clinical pregnancy was achieved in this group of 13 patients. We conclude that in this group of patients, removal of the cumulus prior to insemination may, in some cases, increase the fertilization potential of the oocyte.

The effect of polyploidy on embryo cleavage after in vitro fertilization in humans

The effect of polyploidy on the early development of human embryos is unknown. This study compares the early development of 90 polyploid and 275 diploid human embryos conceived in vitro. Between May 1983 and January 1986, 3081 oocytes were recovered during 631 cycles of laparoscopy for in vitro fertilization (4.9 oocytes/cycle); 1924 oocytes (62.4%) fertilized. There were 90 oocytes with more than two pronuclei (4.7% of fertilized oocytes), identified in 72 cycles (11.4% of cycles). In these cycles, the proportion of diploid oocytes (n = 275) that cleaved (cleavage rate) (92.7%) was significantly greater than the proportion of polyploid oocytes (n = 90) that cleaved (65.5%) (P less than 0.001). The cleavage rate for all diploid oocytes (n = 1834) was 90.4%. There was no significant difference in the stage of development (number of blastomeres; mean +/- standard deviation [SD]) on the day of embryo transfer between diploid (4.3 +/- 2.1) and polyploid (4.1 +/- 2.1) embryos that cleaved, but a plot of the frequency distribution of cleavage stages revealed that significantly more polyploid than diploid embryos had an uneven number of blastomeres at that time (33% versus 8%, respectively; P less than 0.001). Polyploidy confers an immediate developmental disadvantage; one third of polyploid embryos fail to cleave, and those that do divide demonstrate more asynchronous divisions.

Usefulness of serial measurements of serum follicle stimulating hormone, luteinizing hormone and estradiol in patients with premature ovarian failure*

This study tests the hypothesis that serial measurements of serum follicle-stimulating hormone (FSH), luteinizing hormone (LH), and estradiol (E2) are useful in identifying a subset of patients with premature ovarian failure (POF) who may respond to high-dose human menopausal gonadotropin (hMG) therapy. Nineteen patients with POF were studied with weekly measurements of serum FSH, LH, and E2 for five consecutive weeks. Nine patients (group I) showed episodic increases in E2 (greater than 50 pg/ml), seven accompanied by decreases in FSH, and an FSH/LH ratio that was periodically less than 1.0. Ten patients (group II) displayed persistent, nonvarying low E2 and high FSH and LH levels. There was no significant difference in the E2 response to high-dose hMG (48 to 100 ampules hMG/trial) in the two groups, all patients failing to respond. In conclusion, serial assays for FSH, LH, and E2 in patients with POF fail to predict ovarian responsiveness to a trial of high-dose hMG.

Super high estradiol response to gonadotropin stimulation in patients undergoing in vitro fertilization

The likelihood of establishment of a term pregnancy from in vitro fertilization (IVF) is related to the estradiol (E2) pattern and peak level. To examine the influence of super high E2 peak levels (>-2000 pg/ml), we reviewed the clinical outcomes of all IVF cycles with follicular phase E2 levels >2000 pg/ml from May 1982 through June 1987. Among 1651 IVF cycles initiated during this time, 102 cycles (6.2%) had super high E2 levels. Twenty-seven of these cycles occurred in 34 IVF attempts in 12 women. Stimulation was performed with human menopausal gonadotropin (hMG) in 96 cycles and follicle-stimulating hormone (FSH) in 6 cycles. A mean of 9.3±0.7 oocytes per cycle was recovered, of which 5.5±0.5 fertilized and underwent cleavage. In 11 cycles, with a mean of 6.8 oocytes recovered, none fertilized. Polyploid fertilization occurred in 23 of 90 cycles (25.6%), and 40 of 558 fertilized oocytes (7.2%). From these cycles, 10 clinical pregnancies (9.8%) have resulted: 6 pregnancies in 59 cycles with luteal-phase progesterone support (10.2%) and 4 pregnancies in 31 cycles without luteal-phase progesterone support (12.9%). Among the 1549 cycles with peak E2 levels <-2000 pg/ml, 143 (9.2%) resulted in clinical pregnancies. We conclude that there is a small subset of patients who will have super high E2 responses to gonadotropin stimulation and that there is a tendency to stimulate repetitively in this fashion. With regard to pregnancy outcome (1) pregnancy rates are not elevated in this sub-group of high E2 responses, and (2) there does not appear to be an advantage of luteal-phase progesterone support in these super high E2 cycles.

Serum prolactin response to embryo transfer during human in vitro fertilization and embryo transfer

Serum prolactin (PRL) concentrations around the time of embryo transfer (ET) have not been studied, despite the fact that transient hyperprolactinemia regularly occurs in response to laparoscopy for oocyte recovery and ET itself may be stressful enough to induce a PRL rise. Hyperprolactinemia might compromise luteal support for implantation and contribute to the limited success of ET. We measured serum PRL concentrations in 10 normoprolactinemic women immediately before, during, and after ET and compared the PRL response around ET to that induced by laparoscopy as a measure of the competency of the stress-prolactin axis. Nine of ten patients demonstrated a significant PRL response to surgery. The mean (±SEM) intraoperative PRL concentration (124.0±19.6 ng/ml) was significantly higher than the preoperative level (12.3±2.4 ng/ml) (P<0.01). Three hours after surgery PRL levels had decreased (44.8±11.5 ng/ml) but remained above baseline. All subjects were normoprolactinemic 48 hr after laparoscopy. Serum PRL concentration did not change significantly in response to ET, with levels of 10.4±1.7, 12.4±1.1, and 10.6±1.8 ng/ml immediately before, during, and 3 hr after ET, respectively. While laparoscopy for in vitro fertilizationembryo transfer commonly induces hyperprolactinemia, the PRL rise is transient, with no carryover to the time of ET. Embryo transfer itself does not induce a significant PRL rise.