- Email: [email protected]
Advanced semen analysis: a simple screening test to predict intrauterine insemination success
FERTILITY AND STERILITYt VOL. 71, NO. 3, MARCH 1999 Copyright ©1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
Advanced semen analysis: a simple screening test to predict intrauterine insemination success Emmett F. Branigan, M.D.,* M. Antoinette Estes, B.S.,* and Charles H. Muller, Ph.D.† Bellingham In Vitro Fertilization and Infertility Center, Bellingham, and Department of Urology, University of Washington School of Medicine, Seattle, Washington
Received June 9, 1998; revised and accepted November 10, 1998. Presented at the 46th Annual Meeting of the Pacific Coast Fertility Society, Indian Wells, California, April 22–26, 1998. Reprint requests: Emmett F. Branigan, M.D., 2980 Squalicum, Parkway 103, Bellingham, Washington 98225 (FAX: 360-7158126). * Bellingham In Vitro Fertilization and Infertility Center. † Department of Urology, University of Washington School of Medicine. 0015-0282/99/$20.00 PII S0015-0282(98)00503-2
Objective: To determine if a simple screening test of sperm recovery through a density gradient preparation and sperm survival after a 24-hour incubation is predictive of IUI success. Design: Prospective nonrandomized descriptive study. Setting: Tertiary assisted reproductive technology center. Patient(s): Four hundred fourteen couples undergoing IUI for male factor and unexplained infertility. Intervention(s): An advanced semen analysis, which consisted of a basic semen analysis, sperm processing through a density gradient preparation, and a 24-hour sperm incubation, was performed on all couples before beginning IUIs. Main Outcome Measure(s): Cumulative and per cycle pregnancy rates (PRs) were calculated for routine semen parameters, number of sperm processed through density gradient, and percent motile sperm after a 24-hour incubation. Result(s): None of the basic semen analysis parameters accurately predicted IUI success. When the processed total motile sperm available for insemination was $10 3 106 and their 24-hour survival was $70%, 89% (162 of 182) of couples achieved a pregnancy with a 21.4% (162 of 757) per cycle PR compared to a 2.8% (11 of 403) per cycle PR and 4.7% total PR when survival was ,70%. With use of these cutoff values for the advanced semen analysis, the test had a sensitivity of 94% and specificity of 86%. Conclusion(s): The number of motile sperm available for insemination and especially their 24-hour survival are highly predictive of IUI success. This advanced semen analysis is an excellent screening test to evaluate couples considering IUI. (Fertil Sterilt 1999;71:547–51. ©1999 by American Society for Reproductive Medicine.) Key Words: Advanced semen analysis, density gradient, sperm survival, screening test, intrauterine insemination
Intrauterine insemination offers a simple, less expensive, and more acceptable treatment for male factor problems than other advanced reproductive technology procedures. However, the usefulness of IUIs for treating male factor problems remains controversial. The problem is that there is no test or set of tests that is predictive of those likely to benefit from IUIs. Basic semen analysis and even advanced sperm function testing are poor predictors of IUI success, unless severe problems are present (1, 2). Because there are no screening tests to evaluate those likely to benefit from IUIs, most doctors and clinics will often use an empiric trial of three to six or more IUI cycles before moving on to other forms of treatment (3–5).
As a result of this, the patients treated using IUIs are a heterogeneous group. Differences in sperm preparations used, insemination techniques, and the use of ovulation induction agents also make studies comparing the effectiveness of IUIs difficult. We developed a potential screening test we call an “Advanced Semen Analysis.” This test includes a basic semen analysis, followed by a density gradient sperm separation to determine the processed motile sperm fraction and a 24hour incubation to assess sperm survival. This test measures more dynamic aspects of sperm function than the basic semen analysis. Two important physiologic properties of sperm necessary for insemination success are the number 547
of highly motile sperm in a sample and the longevity or survivability of those sperm. Several studies showed that total motile sperm are an important factor for IUI success (1, 5, 6). Processed total motile sperm through a density gradient preparation selects a sperm population with a better motility and morphology index than calculated total motile sperm counts (3, 7). However, because the time between when the insemination is performed relative to oocyte release is variable, sperm survival may be a more important characteristic.
tific, Santa Ana, CA) was used in place of Percoll. The samples were centrifuged at 350 3 g for 15 minutes. The supernatant was removed from each tube, and the pellets were combined with 2 mL of sperm wash (0.5% human albumin 100 in human tubal fluid medium, both from Irvine Scientific) and centrifuged at 250 3 g for 8 minutes. The supernatant was removed, and another wash with 2 mL of sperm wash medium at 250 3 g for 5 minutes was performed. The final pellet was resuspended in 2 mL of sperm wash medium.
The objective of this study is to determine if sperm recovery through a density gradient and sperm survival after a 24-hour incubation is predictive of IUI success in couples with male factor or unexplained infertility.
A sperm count and assessment of sperm motility was performed. The specimen concentration was adjusted to a maximum of 10 3 106 motile sperm per milliliter of medium and incubated for 24 hours at 37°C in 5% CO2; the percentage of sperm motility was reevaluated 24 hours later. All findings were recorded.
MATERIALS AND METHODS
Intrauterine Insemination Cycles
Patient Population Patients who were candidates for IUI for either male factor or unexplained infertility at Bellingham IVF and Infertility Center from November 1994 through November 1997 were included in the study. Inclusion criteria included the following: [1] female age 36 years or younger; [2] documented patent tubes by either hysterosalpingogram or laparoscopy; [3] documented ovulatory cycles with adequate luteal phases either naturally or with clomiphene citrate (no gonadotropin stimulated cycles were included in the study); and [4] men had to have a minimum of 1 3 106 processed total motile sperm. Couples were placed into two groups: those with male factor infertility or with unexplained infertility. The men in the male factor group had either a sperm concentration of ,20 3 106/ml, overall motility of ,50%, or sperm morphology of ,30% by World Health Organization (WHO) criteria (8), or a combination of these abnormalities. The men in the unexplained infertility group had normal semen analysis by these criteria. Informed consent was obtained from all subjects, and the study protocol was approved by the institutional review board of Bellingham IVF and Infertility Center.
Advanced Semen Analysis An advanced semen analysis was obtained from all couples within 2– 6 weeks before starting IUIs. Before collecting the semen sample, men were instructed to abstain from ejaculation for 2–5 days. Samples were obtained by masturbation in a sterile specimen container (Baxter Hospital Supply, Deerfield, IL). The specimens were allowed to liquefy, and a standard WHO (8) criteria semen analysis was performed for semen volume, sperm concentration, motility, sperm morphology, and seminal leukocytes. A discontinuous two-layered Percoll gradient (40%–80%) Percoll (Sigma Chemical Company, St. Louis, MO) was set up in two blue-capped conical Falcon tubes (Baxter Hospital Supply). Beginning in January 1997, Isolate (Irvine Scien548
Branigan et al.
Screening test for IUI success
All insemination cycles were monitored in the follicular phase with use of transvaginal ultrasound (Siemens Sonoline SL-250 7.5-mHz transducer [Siemens Medical Systems, Issaquah, WA]) beginning on cycle days 10 –12 and repeated as needed. When the mean diameter of the lead follicle was .18 mm, 10,000 IU hCG was given IM, and a single IUI was performed 24 –30 hours later. The same sperm preparation as the Advanced Semen Analysis was used with Percoll before January 1997 and Isolate after. The sperm pellet was resuspended in 0.4 mL of sperm wash medium and drawn into an insemination catheter (Tomcat, Provet [Sherwood Medical, St. Louis, MO]) and then deposited high in the fundus of the uterus. The patient remained supine for 15 minutes after IUI. Serum hCG levels were drawn to confirm pregnancies, and 7-week gestation ultrasounds confirmed clinical pregnancies.
Statistical Analysis
x2 analysis was used to compare clinical pregnancy rates (PRs) with the various basic semen analysis parameters and the advanced semen analysis parameters. Linear regression analysis was used to correlate sperm recovered through the density gradient and 24-hour sperm survival with clinical PRs. Receiver operating characteristic curve analysis was used to generate cutoff values for processed sperm counts and 24-hour survival and to determine the predictive power of various sperm parameters. Sensitivity, specificity, and predictive values for positive and negative tests were calculated from these cutoff values.
RESULTS Four hundred fourteen couples completed 1,160 IUI cycles (mean, 2.8 cycles; range, 1–9 cycles). The female age range was from 21 years to 36 years (mean, 33.1 years). The overall PR for all couples was 172 of 414 for a 41.5% rate, and the per cycle PR was 172 of 1,160 or 14.8%. Eighty-two percent (339 of 414) of the men in the study was in the male Vol. 71, No. 3, March 1999
TABLE 1 Distribution of processed total motile sperm counts and 24-hour sperm survival. Processed total motile sperm No. of sperm recovered (3106) ,5 5–10 .10
No. of men with indicated sperm count/total number of men (%)
Pregnancy rate per cycle (%)
33/414 (7.9) 39/414 (9.4) 342/414 (82.6)
0 5.7 16.9*
24-Hour sperm survival
Motility (%) ,20 21–70 .70
No. of men with indicated motility/total number of men (%) 150/414 (36) 66/414 (16) 198/414 (48)
Pregnancy rate per cycle (%) 2.1 3.6 21.4*
* P,.001 versus other groups by x2 analysis.
factor group, and 48% (164 of 339) of them was able to produce a pregnancy. In the unexplained infertility group (8 of 75), only 10% did. No significant relationships were found between PRs and any of the basic semen analysis parameters of concentration, motility, or morphology by x2 analysis using WHO values for normal and abnormal. The receiver operating characteristic analysis was also unable to predict IUI success from individual semen parameters. The distribution of processed total motile sperm counts and 24-hour sperm survival from the advanced semen analysis and their associated PR are shown in Table 1. Linear regression analysis showed a strong positive correlation between processed total motile counts and PR (r 5 0.78; P,.001) and between 24-hour sperm survival and PR (r 5 0.89; P,.001). Significant differences in the PR were noted only in the processed total motile counts of .10 3 106 and 24-hour survival of .70%. Receiver operating characteristic curves suggested cutoff values for total counts of 10 3 106 and 24-hour sperm survival of 70% or more. Using the processed total motile count of 10 3 106 alone was not helpful in predicting those likely to benefit from IUIs because 82.8% of men were above this level. Significant differences in PR were seen only below 10 3 106 level by x2 analysis. Processed total motile sperm count is most useful in setting a lower limit of 5 3 106 (we had no pregnancies below this level) and an optimum level of 10 3 106 or greater. The sperm survival test, however, was highly predictive. When the survival was .70%, 89% (162 of 182) of the couples became pregnant with a 21.4% (162 of 757) per FERTILITY & STERILITYt
cycle PR and accounted for 92% of all the pregnancies in the study. If the survival was ,70%, the per cycle PR was 2.8% (11 of 403), and the cumulative PR was only 4.7%. Of the men with ,70% survival, more than a third (38%) had survivals of ,20%, and their per cycle PR was 2.7%. This poor sperm survival could not be predicted by any semen parameter. The cutoff values of 10 3 106 processed total motile sperm and 24-hour sperm survival of .70% together were used to calculate the sensitivity, specificity, and predictive values of positive and negative tests for the advanced semen analysis. Sensitivity was 94% (162 of 162 1 11 3 100%) with a specificity of 86% (208 of 208 1 33 3 100%). The predictive value of a positive test was 83% (162 of 162 1 33 3 100%), and the predictive value for a negative test was 95% (208 of 11 1 208 3 100%). We also compared men with normal semen analysis (unexplained infertility) with those with abnormal ones (male factor). Twenty-three percent of men with a normal semen analysis failed to meet either one or both of the advanced semen analysis cutoff values. The advanced semen analysis is predictive, independent of whether the basic semen parameters are normal or abnormal.
DISCUSSION These data show that the combination of number of motile sperm available for insemination and their 24-hour survival are highly predictive of IUI success. Twenty-fourhour sperm survival appears to be an especially important sperm characteristic for insemination success. The advanced semen analysis, which measures both of these sperm qualities, would make an excellent screening test to predict those likely to benefit from IUI. The test is easy to perform, requires no specialized laboratory equipment, is cost-effective, and accurately predicts those who could benefit from IUI in men with high sensitivity and specificity. In this study we found that none of the basic semen parameters of concentration, motility, or morphology was related to IUI success. Although processed total motile sperm showed an excellent correlation with PR, it did not screen out many men, because most (83%) had counts of .10 million. Huang et al. (6) found similar results using total motile counts, where PRs remained constant after 5 3 106 motile sperm were reached. Men with the most severe sperm problems can be identified by the basic semen analysis parameters and are not candidates for IUIs. We did not have any pregnancies with processed total motile counts of ,5 3 106, although others have reported pregnancies with lower counts. Our data suggest that processed total motile sperm is important only when it is below this threshold. Twenty-four-hour sperm survival is a sperm characteristic 549
that is not usually tested. Marked differences in sperm survival were noted when reporting the results of standard hamster sperm penetration assays. This sperm function test requires a 24-hour sperm incubation to allow capacitation and acrosome reaction to take place before placing the sperm with the hamster oocytes. A clinical observation that men with poor sperm survival did poorly in inseminations led to the idea for this study. Embryologists frequently observe this differential sperm survival when performing fertilization checks in IVF. Two studies have looked at sperm survival and IVF success. Fuse (9) looked at sperm survival in oligospermic men at several time intervals from 6 to 48 hours of incubation in men doing IVF. He found that if the sperm survived at all after 36 hours, there was a significant difference in fertilization rates in those surviving versus those that did not. Stovall et al. (10) looked at swim-up total motile sperm and 24-hour sperm survival to predict IVF success. They concluded that the number of sperm recovered after swim-up and the percentage that maintain their motility after 24 hours were both helpful in assessing IVF success. No studies have looked at sperm survival with respect to IUI. Sperm survival should be more important in IUIs than in IVF. The amount of time that the sperm need to be functional for the IUI would be much longer than IVF because of the highly variable time period between insemination and oocyte release. In IVF the sperm are placed directly on the oocyte and can be removed in 1–2 hours without affecting fertilization rate. In our program men who had poor overnight motility and who failed to conceive with IUIs have done well in IVF with average or above average fertilization and PRs (unpublished data). This study did not include any cycles in which superovulation was used. Several studies (11–13) showed that the use of gonadotropins is an independent variable that usually improves IUI per cycle PRs by increasing the number of fertilizable oocytes. However, because women have variable responses to gonadotropins, this would introduce another variable into the IUI success. Our goal was to isolate sperm parameters as the only variable for success. We achieved high per cycle PR without the multiple gestation risk, ovarian hyperstimulation, and expense of gonadotropin therapy. In our practice we use the same sperm preparation, ultrasound monitoring, hCG timing, and insemination technique for our therapeutic donor insemination (TDI) cycles (unpublished data). We have a 19.7% per cycle PR in our TDI cycles. This is not statistically significant compared with the 21.3% per cycle PR we achieved in those with .10 million processed total motile sperm and .70% 24-hour survival in the advanced semen analysis. Several studies concluded that IUI is not an effective treatment in “subfertile” men (14 –17). Eighty-three percent of our couples had male factor problems as defined by one, 550
Branigan et al.
Screening test for IUI success
two, or three abnormalities by WHO standards in their basic semen analysis parameters. They had a per cycle PR of 17.8% and a 48% cumulative PR. Twenty-three percent of men with completely normal semen parameters did not meet the cutoff values for 24-hour motility or processed total motile sperm. Their per cycle PR was 1.8%. The advanced semen analysis accurately predicted these “occult” male factors. This study used only one type of sperm preparation, insemination technique, and follicular monitoring technique for insemination timing. We were able to define clear cutoff values for IUI success. Other programs will need to confirm these findings. Different sperm preparations, such as washed and swim-ups, may yield different results. The advanced semen analysis in our hands is an effective screening tool for IUI success. In conclusion, our study suggests that recovery of processed sperm and their overnight survival would make an excellent screening test in the evaluation of couples considering IUI. References 1. Brasch JG, Rawlins R, Tarchala S, Radwanska E. The relationship between total motile sperm count and the success of intrauterine insemination. Fertil Steril 1994;62:150 – 4. 2. Dimarzo SJ, Kennedy JF, Young PE, Herbert SA, Rosenburg DC, Villaneuva B. Effect of controlled ovarian hyperstimulation on pregnancy rates in intrauterine insemination. Am J Obstet Gynecol 1992; 166:1607–13. 3. Allen NC, Herbert CM III, Maxson WS, Rogers BJ, Diamond MP, Wentz AC. Intrauterine insemination: a critical review. Fertil Steril 1985;44:569 – 80. 4. Francavilla F, Romano R, Santucci R, Poccia G. Effect of sperm morphology and motile sperm count on outcome of intrauterine insemination in oligozoospermia and/or asthenozoospermia. Fertil Steril 1990;53:892–7. 5. Kerin JFP, Peek J, Warnes GM, Kirby C, Jeffrey R, Matthews CD, et al. Improved conception rates after intrauterine insemination of washed spermatozoa from men with poor quality semen. Lancet 1994;310: 533– 4. 6. Huang HY, Lee CL, Lai YM, Chang MY, Wang HS, Chang SY, et al. The impact of the total motile count on the success of intrauterine insemination with husband’s spermatozoa. J Assist Reprod Genet 1996; 13:56 – 63. 7. Hoing LM, Devroey P, Van Steirteghem AC. Treatment of infertility because of oligo-astheno-teratozoospermia by transcervical intrauterine insemination of motile spermatozoa. Fertil Steril 1986;45:388 –91. 8. World Health Organization. Laboratory manual for the examination of human semen and sperm-cervical mucus interaction. 3rd ed. New York: Cambridge University Press, 1992:43– 4. 9. Fuse M. Sperm survival test assessing the change of sperm motility after long-term incubation. Nippon Sanka Fujinka Gakkar Zasski 1990; 42:1678 – 84. 10. Stovall DW, Guzick DS, Berga SL, Krasnow JS, Zeleznick AJ. Sperm recovery and survival: two tests that predict in vitro fertilization outcome. Fertil Steril 1994;62:1244 –9. 11. Dodson WC, Haney AF. Controlled ovarian hyperstimulation and intrauterine insemination for treatment of infertility. Fertil Steril 1991; 55:457– 67. 12. Bachus KE, Walmer DK. Superovulation and washed intra-uterine insemination. Contemp Obstet Gynecol 1992;77–90. 13. Cruz RI, Kemmann E, Brandeis VT, Becker KA, Beck M, Beardsley M, et al. A prospective study of intrauterine insemination of processed sperm from men with oligoasthenospermia in superovulated women. Fertil Steril 1986;46:673–7. 14. Ho P-C, Poon IML, Chan SYW, Wang C. Intrauterine insemination is not useful in oligoasthenospermia. Fertil Steril 1989;51:682– 4. 15. Kirby CA, Flakerty SP, Godfrey DM, Warnes GM, Matthews CD. A prospective trial of intrauterine insemination of motile spermatozoa verses timed intercourse. Fertil Steril 1991;56:102–7. 16. Martinez AR, Benardus RE, Voorhorst FJ, Vermeiden JPW, Schoe-
Vol. 71, No. 3, March 1999
maker J. Intrauterine insemination does and clomiphene citrate does not improve fecundity in couples with infertility due to male or idiopathic factors: a prospective, randomized controlled study. Fertil Steril 1990; 53:847–53.
FERTILITY & STERILITYt
17. Ho P-C, So W-K, Chan Y-F, Yeung WS-B. Intrauterine insemination after ovarian stimulation as a treatment for subfertility because of subnormal semen: a prospective randomized controlled trial. Fertil Steril 1992;58:995–9.
551
Advanced semen analysis: a simple screening test to predict intrauterine insemination success Emmett F. Branigan, M.D.,* M. Antoinette Estes, B.S.,* and Charles H. Muller, Ph.D.† Bellingham In Vitro Fertilization and Infertility Center, Bellingham, and Department of Urology, University of Washington School of Medicine, Seattle, Washington
Received June 9, 1998; revised and accepted November 10, 1998. Presented at the 46th Annual Meeting of the Pacific Coast Fertility Society, Indian Wells, California, April 22–26, 1998. Reprint requests: Emmett F. Branigan, M.D., 2980 Squalicum, Parkway 103, Bellingham, Washington 98225 (FAX: 360-7158126). * Bellingham In Vitro Fertilization and Infertility Center. † Department of Urology, University of Washington School of Medicine. 0015-0282/99/$20.00 PII S0015-0282(98)00503-2
Objective: To determine if a simple screening test of sperm recovery through a density gradient preparation and sperm survival after a 24-hour incubation is predictive of IUI success. Design: Prospective nonrandomized descriptive study. Setting: Tertiary assisted reproductive technology center. Patient(s): Four hundred fourteen couples undergoing IUI for male factor and unexplained infertility. Intervention(s): An advanced semen analysis, which consisted of a basic semen analysis, sperm processing through a density gradient preparation, and a 24-hour sperm incubation, was performed on all couples before beginning IUIs. Main Outcome Measure(s): Cumulative and per cycle pregnancy rates (PRs) were calculated for routine semen parameters, number of sperm processed through density gradient, and percent motile sperm after a 24-hour incubation. Result(s): None of the basic semen analysis parameters accurately predicted IUI success. When the processed total motile sperm available for insemination was $10 3 106 and their 24-hour survival was $70%, 89% (162 of 182) of couples achieved a pregnancy with a 21.4% (162 of 757) per cycle PR compared to a 2.8% (11 of 403) per cycle PR and 4.7% total PR when survival was ,70%. With use of these cutoff values for the advanced semen analysis, the test had a sensitivity of 94% and specificity of 86%. Conclusion(s): The number of motile sperm available for insemination and especially their 24-hour survival are highly predictive of IUI success. This advanced semen analysis is an excellent screening test to evaluate couples considering IUI. (Fertil Sterilt 1999;71:547–51. ©1999 by American Society for Reproductive Medicine.) Key Words: Advanced semen analysis, density gradient, sperm survival, screening test, intrauterine insemination
Intrauterine insemination offers a simple, less expensive, and more acceptable treatment for male factor problems than other advanced reproductive technology procedures. However, the usefulness of IUIs for treating male factor problems remains controversial. The problem is that there is no test or set of tests that is predictive of those likely to benefit from IUIs. Basic semen analysis and even advanced sperm function testing are poor predictors of IUI success, unless severe problems are present (1, 2). Because there are no screening tests to evaluate those likely to benefit from IUIs, most doctors and clinics will often use an empiric trial of three to six or more IUI cycles before moving on to other forms of treatment (3–5).
As a result of this, the patients treated using IUIs are a heterogeneous group. Differences in sperm preparations used, insemination techniques, and the use of ovulation induction agents also make studies comparing the effectiveness of IUIs difficult. We developed a potential screening test we call an “Advanced Semen Analysis.” This test includes a basic semen analysis, followed by a density gradient sperm separation to determine the processed motile sperm fraction and a 24hour incubation to assess sperm survival. This test measures more dynamic aspects of sperm function than the basic semen analysis. Two important physiologic properties of sperm necessary for insemination success are the number 547
of highly motile sperm in a sample and the longevity or survivability of those sperm. Several studies showed that total motile sperm are an important factor for IUI success (1, 5, 6). Processed total motile sperm through a density gradient preparation selects a sperm population with a better motility and morphology index than calculated total motile sperm counts (3, 7). However, because the time between when the insemination is performed relative to oocyte release is variable, sperm survival may be a more important characteristic.
tific, Santa Ana, CA) was used in place of Percoll. The samples were centrifuged at 350 3 g for 15 minutes. The supernatant was removed from each tube, and the pellets were combined with 2 mL of sperm wash (0.5% human albumin 100 in human tubal fluid medium, both from Irvine Scientific) and centrifuged at 250 3 g for 8 minutes. The supernatant was removed, and another wash with 2 mL of sperm wash medium at 250 3 g for 5 minutes was performed. The final pellet was resuspended in 2 mL of sperm wash medium.
The objective of this study is to determine if sperm recovery through a density gradient and sperm survival after a 24-hour incubation is predictive of IUI success in couples with male factor or unexplained infertility.
A sperm count and assessment of sperm motility was performed. The specimen concentration was adjusted to a maximum of 10 3 106 motile sperm per milliliter of medium and incubated for 24 hours at 37°C in 5% CO2; the percentage of sperm motility was reevaluated 24 hours later. All findings were recorded.
MATERIALS AND METHODS
Intrauterine Insemination Cycles
Patient Population Patients who were candidates for IUI for either male factor or unexplained infertility at Bellingham IVF and Infertility Center from November 1994 through November 1997 were included in the study. Inclusion criteria included the following: [1] female age 36 years or younger; [2] documented patent tubes by either hysterosalpingogram or laparoscopy; [3] documented ovulatory cycles with adequate luteal phases either naturally or with clomiphene citrate (no gonadotropin stimulated cycles were included in the study); and [4] men had to have a minimum of 1 3 106 processed total motile sperm. Couples were placed into two groups: those with male factor infertility or with unexplained infertility. The men in the male factor group had either a sperm concentration of ,20 3 106/ml, overall motility of ,50%, or sperm morphology of ,30% by World Health Organization (WHO) criteria (8), or a combination of these abnormalities. The men in the unexplained infertility group had normal semen analysis by these criteria. Informed consent was obtained from all subjects, and the study protocol was approved by the institutional review board of Bellingham IVF and Infertility Center.
Advanced Semen Analysis An advanced semen analysis was obtained from all couples within 2– 6 weeks before starting IUIs. Before collecting the semen sample, men were instructed to abstain from ejaculation for 2–5 days. Samples were obtained by masturbation in a sterile specimen container (Baxter Hospital Supply, Deerfield, IL). The specimens were allowed to liquefy, and a standard WHO (8) criteria semen analysis was performed for semen volume, sperm concentration, motility, sperm morphology, and seminal leukocytes. A discontinuous two-layered Percoll gradient (40%–80%) Percoll (Sigma Chemical Company, St. Louis, MO) was set up in two blue-capped conical Falcon tubes (Baxter Hospital Supply). Beginning in January 1997, Isolate (Irvine Scien548
Branigan et al.
Screening test for IUI success
All insemination cycles were monitored in the follicular phase with use of transvaginal ultrasound (Siemens Sonoline SL-250 7.5-mHz transducer [Siemens Medical Systems, Issaquah, WA]) beginning on cycle days 10 –12 and repeated as needed. When the mean diameter of the lead follicle was .18 mm, 10,000 IU hCG was given IM, and a single IUI was performed 24 –30 hours later. The same sperm preparation as the Advanced Semen Analysis was used with Percoll before January 1997 and Isolate after. The sperm pellet was resuspended in 0.4 mL of sperm wash medium and drawn into an insemination catheter (Tomcat, Provet [Sherwood Medical, St. Louis, MO]) and then deposited high in the fundus of the uterus. The patient remained supine for 15 minutes after IUI. Serum hCG levels were drawn to confirm pregnancies, and 7-week gestation ultrasounds confirmed clinical pregnancies.
Statistical Analysis
x2 analysis was used to compare clinical pregnancy rates (PRs) with the various basic semen analysis parameters and the advanced semen analysis parameters. Linear regression analysis was used to correlate sperm recovered through the density gradient and 24-hour sperm survival with clinical PRs. Receiver operating characteristic curve analysis was used to generate cutoff values for processed sperm counts and 24-hour survival and to determine the predictive power of various sperm parameters. Sensitivity, specificity, and predictive values for positive and negative tests were calculated from these cutoff values.
RESULTS Four hundred fourteen couples completed 1,160 IUI cycles (mean, 2.8 cycles; range, 1–9 cycles). The female age range was from 21 years to 36 years (mean, 33.1 years). The overall PR for all couples was 172 of 414 for a 41.5% rate, and the per cycle PR was 172 of 1,160 or 14.8%. Eighty-two percent (339 of 414) of the men in the study was in the male Vol. 71, No. 3, March 1999
TABLE 1 Distribution of processed total motile sperm counts and 24-hour sperm survival. Processed total motile sperm No. of sperm recovered (3106) ,5 5–10 .10
No. of men with indicated sperm count/total number of men (%)
Pregnancy rate per cycle (%)
33/414 (7.9) 39/414 (9.4) 342/414 (82.6)
0 5.7 16.9*
24-Hour sperm survival
Motility (%) ,20 21–70 .70
No. of men with indicated motility/total number of men (%) 150/414 (36) 66/414 (16) 198/414 (48)
Pregnancy rate per cycle (%) 2.1 3.6 21.4*
* P,.001 versus other groups by x2 analysis.
factor group, and 48% (164 of 339) of them was able to produce a pregnancy. In the unexplained infertility group (8 of 75), only 10% did. No significant relationships were found between PRs and any of the basic semen analysis parameters of concentration, motility, or morphology by x2 analysis using WHO values for normal and abnormal. The receiver operating characteristic analysis was also unable to predict IUI success from individual semen parameters. The distribution of processed total motile sperm counts and 24-hour sperm survival from the advanced semen analysis and their associated PR are shown in Table 1. Linear regression analysis showed a strong positive correlation between processed total motile counts and PR (r 5 0.78; P,.001) and between 24-hour sperm survival and PR (r 5 0.89; P,.001). Significant differences in the PR were noted only in the processed total motile counts of .10 3 106 and 24-hour survival of .70%. Receiver operating characteristic curves suggested cutoff values for total counts of 10 3 106 and 24-hour sperm survival of 70% or more. Using the processed total motile count of 10 3 106 alone was not helpful in predicting those likely to benefit from IUIs because 82.8% of men were above this level. Significant differences in PR were seen only below 10 3 106 level by x2 analysis. Processed total motile sperm count is most useful in setting a lower limit of 5 3 106 (we had no pregnancies below this level) and an optimum level of 10 3 106 or greater. The sperm survival test, however, was highly predictive. When the survival was .70%, 89% (162 of 182) of the couples became pregnant with a 21.4% (162 of 757) per FERTILITY & STERILITYt
cycle PR and accounted for 92% of all the pregnancies in the study. If the survival was ,70%, the per cycle PR was 2.8% (11 of 403), and the cumulative PR was only 4.7%. Of the men with ,70% survival, more than a third (38%) had survivals of ,20%, and their per cycle PR was 2.7%. This poor sperm survival could not be predicted by any semen parameter. The cutoff values of 10 3 106 processed total motile sperm and 24-hour sperm survival of .70% together were used to calculate the sensitivity, specificity, and predictive values of positive and negative tests for the advanced semen analysis. Sensitivity was 94% (162 of 162 1 11 3 100%) with a specificity of 86% (208 of 208 1 33 3 100%). The predictive value of a positive test was 83% (162 of 162 1 33 3 100%), and the predictive value for a negative test was 95% (208 of 11 1 208 3 100%). We also compared men with normal semen analysis (unexplained infertility) with those with abnormal ones (male factor). Twenty-three percent of men with a normal semen analysis failed to meet either one or both of the advanced semen analysis cutoff values. The advanced semen analysis is predictive, independent of whether the basic semen parameters are normal or abnormal.
DISCUSSION These data show that the combination of number of motile sperm available for insemination and their 24-hour survival are highly predictive of IUI success. Twenty-fourhour sperm survival appears to be an especially important sperm characteristic for insemination success. The advanced semen analysis, which measures both of these sperm qualities, would make an excellent screening test to predict those likely to benefit from IUI. The test is easy to perform, requires no specialized laboratory equipment, is cost-effective, and accurately predicts those who could benefit from IUI in men with high sensitivity and specificity. In this study we found that none of the basic semen parameters of concentration, motility, or morphology was related to IUI success. Although processed total motile sperm showed an excellent correlation with PR, it did not screen out many men, because most (83%) had counts of .10 million. Huang et al. (6) found similar results using total motile counts, where PRs remained constant after 5 3 106 motile sperm were reached. Men with the most severe sperm problems can be identified by the basic semen analysis parameters and are not candidates for IUIs. We did not have any pregnancies with processed total motile counts of ,5 3 106, although others have reported pregnancies with lower counts. Our data suggest that processed total motile sperm is important only when it is below this threshold. Twenty-four-hour sperm survival is a sperm characteristic 549
that is not usually tested. Marked differences in sperm survival were noted when reporting the results of standard hamster sperm penetration assays. This sperm function test requires a 24-hour sperm incubation to allow capacitation and acrosome reaction to take place before placing the sperm with the hamster oocytes. A clinical observation that men with poor sperm survival did poorly in inseminations led to the idea for this study. Embryologists frequently observe this differential sperm survival when performing fertilization checks in IVF. Two studies have looked at sperm survival and IVF success. Fuse (9) looked at sperm survival in oligospermic men at several time intervals from 6 to 48 hours of incubation in men doing IVF. He found that if the sperm survived at all after 36 hours, there was a significant difference in fertilization rates in those surviving versus those that did not. Stovall et al. (10) looked at swim-up total motile sperm and 24-hour sperm survival to predict IVF success. They concluded that the number of sperm recovered after swim-up and the percentage that maintain their motility after 24 hours were both helpful in assessing IVF success. No studies have looked at sperm survival with respect to IUI. Sperm survival should be more important in IUIs than in IVF. The amount of time that the sperm need to be functional for the IUI would be much longer than IVF because of the highly variable time period between insemination and oocyte release. In IVF the sperm are placed directly on the oocyte and can be removed in 1–2 hours without affecting fertilization rate. In our program men who had poor overnight motility and who failed to conceive with IUIs have done well in IVF with average or above average fertilization and PRs (unpublished data). This study did not include any cycles in which superovulation was used. Several studies (11–13) showed that the use of gonadotropins is an independent variable that usually improves IUI per cycle PRs by increasing the number of fertilizable oocytes. However, because women have variable responses to gonadotropins, this would introduce another variable into the IUI success. Our goal was to isolate sperm parameters as the only variable for success. We achieved high per cycle PR without the multiple gestation risk, ovarian hyperstimulation, and expense of gonadotropin therapy. In our practice we use the same sperm preparation, ultrasound monitoring, hCG timing, and insemination technique for our therapeutic donor insemination (TDI) cycles (unpublished data). We have a 19.7% per cycle PR in our TDI cycles. This is not statistically significant compared with the 21.3% per cycle PR we achieved in those with .10 million processed total motile sperm and .70% 24-hour survival in the advanced semen analysis. Several studies concluded that IUI is not an effective treatment in “subfertile” men (14 –17). Eighty-three percent of our couples had male factor problems as defined by one, 550
Branigan et al.
Screening test for IUI success
two, or three abnormalities by WHO standards in their basic semen analysis parameters. They had a per cycle PR of 17.8% and a 48% cumulative PR. Twenty-three percent of men with completely normal semen parameters did not meet the cutoff values for 24-hour motility or processed total motile sperm. Their per cycle PR was 1.8%. The advanced semen analysis accurately predicted these “occult” male factors. This study used only one type of sperm preparation, insemination technique, and follicular monitoring technique for insemination timing. We were able to define clear cutoff values for IUI success. Other programs will need to confirm these findings. Different sperm preparations, such as washed and swim-ups, may yield different results. The advanced semen analysis in our hands is an effective screening tool for IUI success. In conclusion, our study suggests that recovery of processed sperm and their overnight survival would make an excellent screening test in the evaluation of couples considering IUI. References 1. Brasch JG, Rawlins R, Tarchala S, Radwanska E. The relationship between total motile sperm count and the success of intrauterine insemination. Fertil Steril 1994;62:150 – 4. 2. Dimarzo SJ, Kennedy JF, Young PE, Herbert SA, Rosenburg DC, Villaneuva B. Effect of controlled ovarian hyperstimulation on pregnancy rates in intrauterine insemination. Am J Obstet Gynecol 1992; 166:1607–13. 3. Allen NC, Herbert CM III, Maxson WS, Rogers BJ, Diamond MP, Wentz AC. Intrauterine insemination: a critical review. Fertil Steril 1985;44:569 – 80. 4. Francavilla F, Romano R, Santucci R, Poccia G. Effect of sperm morphology and motile sperm count on outcome of intrauterine insemination in oligozoospermia and/or asthenozoospermia. Fertil Steril 1990;53:892–7. 5. Kerin JFP, Peek J, Warnes GM, Kirby C, Jeffrey R, Matthews CD, et al. Improved conception rates after intrauterine insemination of washed spermatozoa from men with poor quality semen. Lancet 1994;310: 533– 4. 6. Huang HY, Lee CL, Lai YM, Chang MY, Wang HS, Chang SY, et al. The impact of the total motile count on the success of intrauterine insemination with husband’s spermatozoa. J Assist Reprod Genet 1996; 13:56 – 63. 7. Hoing LM, Devroey P, Van Steirteghem AC. Treatment of infertility because of oligo-astheno-teratozoospermia by transcervical intrauterine insemination of motile spermatozoa. Fertil Steril 1986;45:388 –91. 8. World Health Organization. Laboratory manual for the examination of human semen and sperm-cervical mucus interaction. 3rd ed. New York: Cambridge University Press, 1992:43– 4. 9. Fuse M. Sperm survival test assessing the change of sperm motility after long-term incubation. Nippon Sanka Fujinka Gakkar Zasski 1990; 42:1678 – 84. 10. Stovall DW, Guzick DS, Berga SL, Krasnow JS, Zeleznick AJ. Sperm recovery and survival: two tests that predict in vitro fertilization outcome. Fertil Steril 1994;62:1244 –9. 11. Dodson WC, Haney AF. Controlled ovarian hyperstimulation and intrauterine insemination for treatment of infertility. Fertil Steril 1991; 55:457– 67. 12. Bachus KE, Walmer DK. Superovulation and washed intra-uterine insemination. Contemp Obstet Gynecol 1992;77–90. 13. Cruz RI, Kemmann E, Brandeis VT, Becker KA, Beck M, Beardsley M, et al. A prospective study of intrauterine insemination of processed sperm from men with oligoasthenospermia in superovulated women. Fertil Steril 1986;46:673–7. 14. Ho P-C, Poon IML, Chan SYW, Wang C. Intrauterine insemination is not useful in oligoasthenospermia. Fertil Steril 1989;51:682– 4. 15. Kirby CA, Flakerty SP, Godfrey DM, Warnes GM, Matthews CD. A prospective trial of intrauterine insemination of motile spermatozoa verses timed intercourse. Fertil Steril 1991;56:102–7. 16. Martinez AR, Benardus RE, Voorhorst FJ, Vermeiden JPW, Schoe-
Vol. 71, No. 3, March 1999
maker J. Intrauterine insemination does and clomiphene citrate does not improve fecundity in couples with infertility due to male or idiopathic factors: a prospective, randomized controlled study. Fertil Steril 1990; 53:847–53.
FERTILITY & STERILITYt
17. Ho P-C, So W-K, Chan Y-F, Yeung WS-B. Intrauterine insemination after ovarian stimulation as a treatment for subfertility because of subnormal semen: a prospective randomized controlled trial. Fertil Steril 1992;58:995–9.
551