The relationship of sperm parameters to cycle fecundity in superovulated women undergoing intrauterine insemination

FERTILITY AND S TERILITY

Vol. 52, No.2, August 1989 Printed in U.S.A.

Copy right "' 1989 The American Fertili ty S ociety

The relationship of sperm parameters to cycle fecundity in superovulated women undergoing intrauterine insemination

Peter M. Horvath, M.D .* Michael Bohrer, M.D. Robert M. Shelden, Ph.D. Ekkehard Kemmann, M.D . Department of Obstetrics and Gy necology, Division of Reproductive Endocrinology, University of M edicine and Dentistry of N ew J ersey Robert W ood Johnson M edical S chool, N ew Brunswick, N ew J ersey

Semen parameters of raw and prepared (post-swim-up) specimens from 451 cycles of intrauterine insemination (lUI) were analyzed in relation to cycle fecundity in 232 patients undergoing ovarian stimulation with sequential clomiphene citrate/ menotropin therapy. Pregnancy occurred in 42 cycles, resulting in an overall pregnancy rate of 17. 7%, and a cycle fecundity of9.3%. Cycle fecundity was positively correlated with the parameters of post-swim-up log sperm density (r = 0.994), and with log total motile sperm inseminated (r = 0.964; inseminates were limited to a maximum of 20 million total motile sperm). Post-swim-up motility did not correlate (r = 0.308) with cycle fecundity; however, most specimens had a motility of > 40 % post-swim-up. Only one pregnancy occurred when less than 1 million motile sperm were inseminated (38 cycles) , which resulted in a cycle fecundity of 2.6% for t hese cycles. This may represent the threshold of eft'e ctiveness for lUI in this setting. Highest cycle fe cundity was obtained with an inseminate containing approximately 10 million or more motile sperm. Parameters of raw samples correlated less well with cycle fecundity than did prepared specimens. Analysis of post-swim-up semen parameters can provide useful prognostic information for women undergoing lUI with ovarian stimulation; this information is helpful in counseling patients regarding their chances of success with this therapy. Fertil Steril 52:288, 1989

The timed application of washed spermatozoa by intrauterine insemination (lUI) has been used in the therapy of a number of infertility situations. 1 The effectiveness of this approach depends on several factors. Reasonable pelvic anatomic integrity and appropriate timing of the insemination in relation to ovulation are obvious prerequisites for success. In this regard, we recently demonstrated2 that active ovulation management further increases the efficacy of lUI. Another major determinant of out-

Received January 6, 1989; revised and accep ted April4, 1989.

* Present address and request for reprints: Peter M . H orvath , M .D ., Department of Obstetrics and Gynecology, Divi sion of Reproductive Endocrinology, Albany M edical Cen ter H ospital, Albany, New York 12208.

288

Horvath et al.

Semen parameters and success in l UI

come is the quantity and quality of the sperm specimen. Indeed, there is some debate 3 •4 about the efficacy of lUI as a treatment for oligospermic or asthenospermic males. Lastly, it should be remembered that protocols for sperm preparation for lUI vary, and thus can influence results. We have been interested in studying the relation between sperm parameters and lUI efficacy in women undergoing ovulation management. A preliminary analysis of 642 treatment cycles5 suggested that lUI with < 1 million total motile sperm was rarely successful in this setting. When patients were inseminated with higher numbers of active sperm, conception rates were similar in all groups, at about 7% per treatment cycle. The current study was undertaken to examine the relation between Fertility and S terility

specific sperm parameters (raw and post-swim-up) and cycle fecundity, specifically, to define if, and to what degree, cycle fecundity depends on number and motility of the inseminate. For the purposes of the current analysis, all data were obtained from a study population in which only one method of ovulation management and lUI timing was used. MATERIALS AND METHODS All patients (n = 232) treated with an ovulation stimulating regimen consisting of sequential clomiphene citrate (CC)/human menopausal gonadotropin (hMG)/human chorionic gonadotropin (hCG)6 who underwent lUI in 1987 were subjects of the study. Nearly all patients underwent diagnostic laparoscopy to assess pelvic status before menotropin stimulation. One hundred forty-five patients had ovulatory dysfunction as their primary diagnosis. These patients included those with polycystic ovarian disease (n = 54), luteal phase defect (n = 43), hypothalamic hypogonadotropic amenorrhea (n = 8), or those with menstrual irregularity, but who did not fit the criteria for inclusion into one of these diagnostic categories. Polycystic ovarian disease was defined either by the presence of altered gonadotropin ratios, classic Stein-LevinthaI phenotype and history, or by ovarian morphology as seen at laparoscopy. Luteal phase defect was diagnosed either by out-of-phase endometrial histology (>2 days) or by suboptimal luteal phase progesterone profiles7 in at least two cycles. The remainder of the patients (with normal ovulatory function) had the following primary diagnoses: male factor (n = 39), cervical factor (n = 14), unexplained infertility (n = 13), and endometriosis (n = 21). Most patients with endometriosis had minimal or mild disease treated at the time of laparoscopic diagnosis with CO 2 laser. Treatment was individualized for patients with more advanced endometriosis, using medical and surgical therapy before menotropin stimulation. Male partners of couples with a primary diagnosis of male factor infertility were referred for urologic consultation before treatment with lUI. Many couples had more than one infertility factor. Most patients with ovulatory dysfunction had been treated unsuccessfully with 6 or more months of CC therapy before menotropin stimulation; patients with normal ovulatory function were treated with CC/hMG/hCG to manage ovulation for IUI. 2 Patient ages ranged from 23 to 45 years (mean, 32.8), and duration of infertility from 1 to 12 years (mean, 3.4). Details ofthe CC/hMG/hCG protocol have been Vol. 52, No.2, August 1989

SEMEN ANALYSIS < 20 t.4ILLlON/ML

NORMAL

<

OR 35" MQTlUlY

1. MALE FACTOR

CERVICAL MUCUS (CM) Spinnbarkeit

>4cm

~Pirmbarkeit <4cm

2. CM DEFICIENCY

POSTCOITAL TEST (PCT) peT SCORE ~ PC~ ;CORE >7 / ~ NORMAL •

3. POOR PCT

Figure 1 Algorithm depicting criteria used in determination of various diagnostic groups as an indication for lUI. Postcoital testing was according to the method of Skaf and Kemmann.8 Note: "normals" were not treated with lUI.

published.6 For reasons of cost efficiency, and to reduce side effects (i.e., multiple gestation), we instituted its use as an intermediate step in the management of ovulation, generally after failure of CC therapy, but before the patient would be treated with hMG/hCG. Briefly, the treatment consists of CC 50 to 100 mg orally, daily from day 3 to 7, followed by hMG 150 to 225 IU, intramuscularly (1M), on alternate days, or daily from day 8 to 9. Serum estradiol (E 2) levels were obtained every other day from day 8 to day 9. When E2 levels reached 400 to 500 pg/ml, hMG stimulation was discontinued, and hCG, 10,000 IU was administered 1M. Vaginal sonography (GE 3600, 5.0 MHz probe, General Electric Corporation, Schenectady, NY) was used in most patients to confirm the presence of at least one preovulatory (~15 mm) follicle just preceding hCG injection. A single transcervical, high-fundal lUI was performed in the semilithotomy position, using a Shepard cannula (Cook Corporation, Bloomington, IN), 28 to 29 hours after hCG administration. Patients were allowed to leave the table immediately after the procedure; no restriction of activity, including intercourse, was recommended. Serum pregnancy testing was performed on patients who remained without menses for 17 days after lUI; two rising titers of {j-hCG were considered evidence of pregnancy occurrence. Intrauterine insemination was offered to couples for male factor infertility, poor cervical mucus, or poor results on postcoital testing; couples were grouped according to indication for lUI following the algorithm in Figure 1. Male factor infertility was defined as present when two or more abnormal semen analyses (density < 20 M/ml; motility < 35% at 2 hours; <50% with normal morphology) performed at least 1 month apart were obtained. If male factor infertility was not present, couples Horvath et al.

Semen parameters and success in lUI

289

underwent preovulatory postcoital testint 8 to 12 hours after intercourse for evaluation of cervical mucus and sperm-mucus interaction. If postcoital testing results were poor, lUI was recommended beginning immediately; borderline postcoital tests, conversely, were repeated in a subsequent cycle be- . fore assigning the couple to lUI therapy. Sperm specimens were collected by masturbation into a sterile container, and processed within 30 to 120 minutes. Ham's F-10 (Gibco, Grand Island, NY) supplemented with 0.5% human serum albumin was used as washing medium. Initial semen analysis was performed on 100 p,l of liquified raw specimen; the remainder was admixed with 2.0 ml of washing medium, and centrifuged at 800 X g for 5 minutes. After removing the supernatant, the pellet was overlaid with 0.5 to 0.8 ml of washing medium. A swim-up was carried out for 45 to 75 minutes at 37°C in 5% CO2-air, with 15 degrees' tube inclination. The supernatant was carefully removed and a 50-p,1 aliquot was taken for post-swimup semen analysis; the remainder was used for insemination, accepting specimens that contained >20 million total motile spermatozoa, which were diluted to 20 million total motile before insemination. If no motile sperm were recovered after washing and swim-up, lUI was not performed. All semen analyses were performed on an automated analysis system (Cellsoft Semen Analysis System, Cryo Resources Ltd., New York, NY), and provided data on volume, density, motility, and total motile count. Parameters for our laboratory for computerized semen analysis have been published. 9 Double-antibody radioimmunoassay was used for determination of both E2 and ,B-hCG (Leeco Diagnostics Inc., Southfield, MI). Intra-assay and interassay variations were all <10%. Data were analyzed for significance using standard statistical methods including chi-square, Student's t-test, and Fisher's exact test where appropriate. Log transformation of data and all statistical calculations used Lotus 1-2-3 software (Lotus Development Corporation, Cambridge, MA). Cycle fecundity was defined as the ratio of pregnancy per treatment cycles, and corresponds to monthly probability of conception as described by Cramer et a1. 10

RESULTS Semen from 232 patients in 466 treatment cycles were prepared for lUI; 15 (3%) specimens provided no motile sperm after preparation, and were ex290

Horvath et al.

Semen parameters and success in lUI

Table 1

Treatment Outcome In Relation To lUI Indication

Male factor Cervical factor Poor postcoital test

No. of subjects

No. of pregnancies

Percentage of pregnancies

175 121 155

6 14

3.4" 11.6 14.2

22

" P < 0.01, compared with other indications, Fisher's exact test.

cluded from analysis. Indications for the use of lUI were male factor infertility in 175 treatment cycles, cervical factor in 121 treatment cycles, and poor postcoital testing in 155 treatment cycles (Table 1). Pregnancy occurred in 42 cycles (2 pregnancies occurred in one patient), which resulted in an overall pregnancy rate of 17.7%, and a cycle fecundity of 9.3% for the study population. There were no differences between pregnant and nonpregnant patients with regard to age, height, weight, number of treatment cycles, or duration of infertility; nor within their cycles with regard to the maximum serum E2 level obtained within 24 hours of hCG administration. Twin and tubal gestations occurred in 2 cycles each (4.8%), and 12 conceptions ended in first trimester miscarriage (29%). The cycle fecundity differed among the various treatment groups (Table 1), and was significantly higher for patients being treated for cervical factor infertility or poor postcoital test than for patients who received lUI for male factor infertility (P < 0.01, Fisher's exact test). Effects of sperm washing and swim-up are illustrated in Figure 2. While sperm preparation increased the percentage of motile sperm in the prepared compared with the raw specimen, (Fig. 2C), sperm density and numbers of total motile sperm were reduced by sperm processing (Fig. 2A, B). This effect of sperm preparation was uniform in all cycles in that similar distributions were obtained in pregnancy and nonpregnancy cycles. Mean values for various semen parameters of the raw and prepared specimens in conception and nonconception cycles are presented in Table 2. All semen parameters except volume were significantly lower (P < 0.01) in patients with male factor infertility, compared with the other two groups; however, mean recovery of total motile sperm after preparation for lUI was equally efficient in male factor samples compared with the other two groups (16% to 18%). Within each group, sperm recovery rates were similar in pregnancy and nonpregnancy Fertility and Sterility

A

...

i •

100

Iii is

10

100

1000

SPERW DENSITY (WIWON/WL)

B

... ~

z

~

jtOO

i!!

en

is

rowo

C

10

100

1000

SPERW (WIWON)

~'50

~

5 '00

! Q

50

w

m

~

~

50

~

~

~

~

~

PERCENT WOTiUTY

Figure 2 The effect of semen preparation on the log·transformed distributions of sperm density and total motile sperm (A and B, respectively), and the distribution of sperm motility (e) in the raw and prepared semen specimens from all cycles.

cycles. Of the pregnancy cycle samples, one sample from the male factor group had exceptionally high recovery, which skewed the mean recovery in this subgroup (47%); removal of this sample from analysis results in a mean recovery of 12%, which is similar to male factor samples that did not result in pregnancy. No differences in sperm parameters were apparent within each group between conception and nonconception cycles, either in the raw or in the post-swim-up specimens. However, when cycles were analyzed without regard to indication for lUI, sperm parameter differences were evident between conception and nonconception cycles (Table 2). While all sperm parameters were increased in the raw specimens of pregnancy cycles compared with nonpregnancy cycles, none of these differencesregardless of log-transformation-achieved statistical significance (Student's t-test). Conversely, log-transformed sperm density, total motile sperm count, and adjusted total motile count (adjusted to Vol. 52, No.2, August 1989

account for a maximum of 20 million total motile in the actual inseminates used) were significantly increased in post-swim-up specimens from conception cycles compared with nonconception cycles (P < 0.05, 0.05, and 0.02, log density, log total motile, and log-adjusted total motile, respectively) . Cycle fecundity generally increased with increasing sperm density, total motile count, and motility (Fig. 3). Sperm density and total motile count of post-swim-up specimens showed better correlation with cycle fecundity than did these parameters in the raw specimens (r = 0.994 and 0.964, postswim-up log density and log total motile, respectively, and r = 0.837 and 0.721, raw specimen log density and log total motile, respectively). While percent motility of the raw specimen correlated with cycle fecundity to some degree (r = 0.792, for motility ranging from 0% to 70%), the motility of the post-swim-up specimen did not correlate with cycle fecundity (r = 0.308); however, most prepared specimens had at least 40% motility. Pregnancy was unlikely if there was less than 40% motility in the post-swim-up specimen: only two pregnancies occurred, with 33% and 34% motility, respectively, and a total motile count of 6.8 million and 14.4 million, respectively. The cycle fecundity for inseminates with sperm motility of <40% was 4.9%. While cycle fecundity was improved (9.7%) in cycles with post-swim-up motility > 40%, it should be noted that sperm preparation was highly successful in obtaining specimens with very high motility (c.80%) in most (67%) cases. Regarding total motile sperm obtained post-swim-up, maximum efficacy appeared to be reached at approximately 10 million sperm (Fig. 3B). We could not demonstrate that insemination with higher numbers of motile sperm was associated with better cycle fecundity. Less than 1 million total motile sperm in the inseminate was essentially ineffective: only one pregnancy resulted (from a specimen containing 0.680 million total motile sperm with 45% motility) out of 38 inseminates containing <1 million total motile sperm. Thus, the cycle fecundity for CCjhMGjhCG using lUI with inseminates of <1 million total motile sperm was 2.6%. Between 1 and 10 million, there was an apparant positive correlation between total motile and cycle fecundity (Fig. 3). Conversely, for treatments with an inseminate containing> 10 million motile sperm, the cycle fecundity was 11.6%. Minimal cramping and spotting was reported by a minority of patients during and immediately following the procedure. In addition, one patient exHorvath et al.

Semen parameters and success in lUI

291

Table 2 Mean (SEM)a Semen Characteristics of Raw and Post-Swim-up Specimens in Conception and Nonconception Cycles Grouped by Indication for Intrauterine Insemination Post-swim-up specimen Raw specimen

Indication

N

Volume

Density

Motility

Total. motile

ml

Mimi

%

M

MImi

43.5 (3.7) 27.5 (9.2) 44.1 (3.9)

17.9 (1.7) 30.8 (19.1) 17.4 (1.6)

Density

Total motile percent recovered

Adjusted total motile·

Motility

Total motile

%

M

26 56 25

72 (1.9) 73 (5.3) 72 (1.9)

7.1 (0.7) 12.8 (8.4) 6.9 (0.7)

16 47 16

5.9 (0.5) 6.4 (2.6) 5.9 (0.5)

Percent recovered

M

MaleFactor All Pregnant Nonpregnant

175 6 169

2.6 (0.1) 3.0(0.6) 2.6 (0.1)

68.4 (4.2) 55.3 (9.6) 68.8 (4.3)

24.7 (1.2) 18.3 (3.6) 25.0 (1.2)

Cervical All Pregnant Nonpregnant

121 14 107

2.7 (0.1) 3.1 (0.4) 2.7 (0.1)

97.4 (0.1) 79.0 (13.0) 99.8 (7.4)

40.9 (1.8) 34.4 (3.6) 41.7 (2.0)

107 75 112

(10.0) (15.2) (11.1)

46.8 (4.2) 43.3 (9.1) 47.3 (4.6)

48 55 47

85 (1.7) 83 (5.4) 85 (1.8)

19.6 (1.8) 17.4 (4.0) 19.9 (1.9)

18 23 18

13.0 (0.7) 13.3 (2.0) 12.9 (0.7)

Poor postcoital test All Pregnant Nonpregnant

155 22 133

2.5 (0.1) 1.9 (0.2) 2.6 (0.1)

108 136 103

(6.5) (17.4) (6.9)

36.1 (1.4) 41.7 (3.5) 35.1 (1.5)

109 120 107

(10.6) (22.2) (11.9)

45.8 (4.5) 49.3 (12.4) 45.2 (4.8)

42 36 44

83 (1.4) 82 (4.0) 83 (1.5)

18.2 (1.9) 21.5 (6.2) 17.7 (1.9)

17 18 17

11.3 (0.6) 12.0 (1.5) 11.2 (0.7)

42

2.5 (0.3)

106

(11.4)

36.0 (2.6)

92.2 (13.8)

44.7 (7.8)

42

81 (2.9)

18.9 (3.7)

20

11.7 (1.1)

409

2.7 (0.1)

32.7 (1.0)

82.6 (5.3)

34.3 (2.2)

39

79 (1.1)

13.9 (0.9)

17

All pregnant All nonpregnant

88.3 (3.6)

a SEM, standard error of the mean. • After dilution of specimens containing high numbers of total motile

perienced an occurrence of clinical salpingitis within 24 hours of lUI, which required antibiotic therapy. DISCUSSION

Our study indicates that, in women undergoing lUI in a specific clinical setting-ovulation management with CC/hMG/hCG-certain semen parameters are correlated with treatment outcome. Specifically, sperm density and total motile count of the post-swim-up specimen showed very high correlations with cycle fecundity. While only one pregnancy occurred when post-swim-up sperm density was <1 million per milliliter, a near-linear relationship existed between cycle fecundity and log sperm density with higher concentrations of sperm (Fig. 3A). Similarly, log total motile count was highly correlated with cycle fecundity, specifically between 1 and 10 million total motile; total motile sperm> 10 million in the inseminate did not increase cycle fecundity further. Post-swim-up sperm motility did not correlate with monthly chance of pregnancy occurrence once a minimum of 40% motility was present in the inseminate. Generally, analysis of the raw specimens provided poorer correlations between cycle fecundity and se292

Horvath et aI.

Semen parameters and success in lUI

9.4 (0.37)

sperm to not more than 20 million total motile .

men parameters than did analysis of the prepared specimens. Clearly, the techniques used in sperm preparation are important in the ultimate success of lUI. Sperm washing and swim-up yielded results consistent with any extraction process (Fig. 2), in that sperm preparation consistently improved motility; however, this gain occurred at the expense of diminished numbers of recovered motile sperm. Average sperm recovery rates were consistant in all diagnostic groups, and were similar to those reported by others using similar preparation methodology.ll Because total numbers of motile sperm used in the actual inseminate correlate well with cycle fecundity, development of better methods of sperm recovery may be useful in maximizing recovery of motile sperm. Although not tested in the present study, one may speculate that alternative semen preparation techniques could improve recovery of motile sperm, and therefore treatment success. For instance, Berger et al. 12 reported a 59% recovery rate using a discontinuous Percoll gradient technique, compared with 18% obtained by standard swim-up. Higher recovery rates (58%) have also been reported by Harris et al./ 3 who used a much smaller semen aliquot (0.05 ml) for swim-up. Fertility and Sterility

-_ A

..........

.15 ....... POIT-IWII.P ..a:I8I

10

B

10C1

SPERM DENSITY (MIWON/ML)

.15

~Z

:>

~

~ ~

u

.1

... 10

C

100

TOMO SPERM (MIWON)

Figure 3 Correlation of log-transformed sperm density and total motile sperm (A andB, respectively), and percent motility (C) with cycle fecundity. The asterisk in (B) illustrates cycle fecundity calculated with total motile sperm count adjusted to a maximum of 20 million motile sperm.

While miscarriage rates did not correlate with numbers of sperm inseminated, the overall miscarriage rate was high, as noted in other studies involving the use of human gonadotropins. 14 The multiple gestation rate, conversely, was low «5%) in this study-which used a CCjhMGjhCG stimulation protocol-compared with the higher rates reported with hMGjhCG and IUI. 15 ,16 By the same token, the cycle fecundity of CCjhMGjhCG therapy tends to be lower than that of hMGjhCG. 2 Thus, there appears to be a tradeoff between safety and efficacy among various methods of ovulation management and lUI. As an additional precaution, our protocol called for a limit of 20 million motile sperm in the inseminate,15 specifically as optimal efficacy of lUI seemed to have been reached at 10 million motile sperm. However, we feel that this limitation is more critical in hMGjhCG therapy, which has a higher degree of superovulation than CCjhMGjhCG stimulation. Vol. 52, No.2, August 1989

Relatively few data are available on the relation between semen parameters and treatment outcome in women undergoing lUI. In a series of 25 couples undergoing insemination for poor postcoital testing (specifically excluding cervical or male factor problems), Arny and Quagliarello 11 examined treatment outcome using logistic discriminant analysis of various semen parameters from raw and post-swim-up samples. In addition to semen parameters, they analyzed the variables of patient age and duration of infertility. In their analysis, only one variable, the post-swim-up motility, provided significant discriminating ability between patients who would (post-swim-up motility> 78.7%) and would not (motility < 78.7%) conceive. While our current data were not evaluated by the same statistical methodology as those of Arny and Quagliarello,l1 our data do not support the concept that post-swim-up motility, as a sole parameter, is a good predictor of success in women receiving lUI. In fact, once a minimum of approximately 40% motility was achieved in the prepared specimen, further improvement in motility did not improve cycle fecundity (Fig. 3C). It should be noted, however, that our process of sperm preparation was highly successful in obtaining specimens with >80% motility. Thus, we found that other post-swim-up semen parameters (density and total motile count) were more highly correlated with pregnancy occurrence than motility. Unfortunately, it is not possible to assess functional capacity of sperm by the parameters of the semen analysis alone. Currently, the heterologous sperm penetration assay has been advocated by many investigators to assess sperm function. However, the sperm penetration assay is expensive, is difficult to reproduce, requires extensive laboratory experience, and has limited value in counseling most couplesP It is unclear whether the newer hemizona assay18 may fulfill the potential as a better in vitro test for sperm function. Notwithstanding the need for reliable, costeffective, yet practical assays for functional capacity of sperm, conventional sperm parameters obtained during the semen preparation for lUI provide a helpful framework within which patients with male factor infertility can be guided regarding the potential of treatment success using lUI. It appears that in situations where post-swim-up specimens yield <1 million total motile sperm, the chance of success with this approach is very low. Therefore, it has been our practice to counsel these patients about alternative therapy such as donor Horvath et al.

Semen parameters and success in lUI

293

artificial insemination. In patients with total motile counts of 1 to 10 million, results are improved, and couples are usually encouraged to continue therapy for up to six cycles; yet counseling is individualized and treatment strategies may depend on the consistency and overall quality of specimens. Patients who are inseminated with > 10 million' motil~ sperm seem to obtain the most benefit from lUI. Our data suggest that valuable prognostic information can be obtained from analysis of lUI specimens in the setting of active ovulation management with CC/hMG/hCG. Generally, inseminates containing a minimum of 1 million total motile sperm with a motility of at least 40% seem to be required for successful pregnancy achievement. Improved success can be expected with specimens containing between 1 and 10 million total motile sperm, reaching maximal efficacy with inseminates containing> 10 million total motile sperm. This information can be used in counseling couples regarding their chances of success with ovulation management and lUI. REFERENCES 1. Allen NC, Herbert CM, III, Maxson WS, Rogers BJ, Diamond MP, Wentz AC: Intrauterine insemination: a critical review. Fertil Steril44:569, 1985 2. Kemmann E, Bohrer M, Shelden R, Fiasconara G, Beardsley L: Active ovulation management increases the monthly probability of pregnancy occurrence in ovulatory women who receive intrauterine insemination. Fertil Steril48:916, 1987 3. Kerin J, Quinn P: Washed intrauterine insemination in the treatment of oligospermic infertility. Sem Reprod Endocrinol5:23, 1987 4. Moghissi KS: Some reflections on intrauterine insemination. Fertil Steril46:13, 1986 5. Bohrer M, Kemmann E, Pasquale S, Shelden RM: The significance of the total number of motile sperm delivered with intrauterine insemination (lUI) in menotropin treated women. (Abstr. P-159) Presented at the Forty-Second Annual Meeting of The American Fertility Society and the

294

Horvath et al.

Semen parameters and success in lUI

Eighteenth Annual Meeting of The Canadian Fertility and Andrology Society, Toronto, Ontario, Canada, September 27 to October 2, 1986, Published by The American Fertility Society in program supplement, 1986, P 56 6. Kemmann E, Jones JR: Sequential clomiphene citratemenotropin therapy for induction or enhancement of ovulation. Fertil Steril39:772, 1983 7. Abraham GE, Maroulis GB, Marshall JR: Evaluation of ovulation and corpus luteum function using measurements of plasma progesterone. Obstet Gynecol44:522, 1974 8. Skaf RA, Kemmann E: Postcoital testing in women during menotropin therapy. Fertil Steril37:514, 1982 9. Horvath PM, Beck M, Bohrer MK, Shelden RM, Kemmann E: A prospective study on the lack of development of antisperm antibodies in women undergoing intrauterine insemination. Am J Obstet Gynecol160:631, 1989 10. Cramer DW, Walker AM, Schiff I: Statistical methods in evaluating the outcome of infertility therapy. Fertil Steril 32:80,1979 11. Amy M, Quagliarello J: Semen quality before and after processing by a swim-up method: relationship to outcome of intrauterine insemination. Fertil Steril48:643, 1987 12. Berger T, Marrs RP, Moyer DL: Comparison oftechniques for selection of motile spermatozoa. Fertil Steril 43:268, 1985 13. Harris SJ, Milligan MP, Masson GM, Dennis KJ: Improved separation of motile sperm in asthenospermia and its application to artificial insemination homologous (AIH). Fertil Steril36:219, 1981 14. Bohrer M, Kemmann E: Risk factors for spontaneous abortion in menotropin-treated women. Fertil Steril 48:571, 1987 15. Shelden R, Kemmann E, Bohrer M, Pasquale S: Multiple gestation is associated with the use of high sperm numbers in the intrauterine insemination specimen in women undergoing gonadotropin stimulation. Fertil Steril 49:607, 1988 16. Dodson WC, Hughes CL, Haney AF: Multiple pregnancies conceived with intrauterine insemination during superovulation: an evaluation of clinical characteristics and monitored parameters of conception cycles. Am J Obstet Gynecol 159:382, 1988 17. Mao C, Grimes DA: The sperm penetration assay: can it discriminate between fertile and infertile men? Am J Obstet Gynecol159:279, 1988 18. Burkman LJ, Coddington CC, Franken DR, Kruger TF, Rosenwaks Z, Hodgen GD: The hemizona assay (HZA): development of a diagnostic test for the binding of human spermatozoa to the human hemizona pellucida to predict fertilization potential. Fertil Steril 49:688, 1988

Fertility and Sterility