COMPARISON OF FOUR METHODS FOR SPERM

PREPARATION FOR IUI

S.-S. REN, G.-H. SUN, C.-H. KU, D.-C. CHEN, and G.-J. WU

National Defense Medical Center, National Defense University,Taipei, Taiwan, R.O.C.

We compared the pregnancy rates and sperm characteristics of semen prepared by the albumin,

Percoll, Puresperm, and swim-up methods. Semen analysis was performed by a computer-assisted

system (CASA). The overall pregnancy rate was 14% per patient and 11% per cycle. The albumin

and Percoll methods had the highest pregnancy rates, up to 12% per cycle. We then compared each

method with the albumin method. The swim-up technique yielded the highest percentage of motile

sperm. The Percoll method yielded statistically significant changes in average path velocity and

straight-line velocity, while the Puresperm method revealed the same statistical changes in total

concentration and hyperactive motile sperm percentage. After adjusting for age, methods, and

CASA estimates, only straight-line velocity was significantly correlated with pregnancy. Although

the Percoll method was associated with the best pregnancy rate, this product has been withdrawn

from the market because of toxic contamination. Puresperm had replaced it in our laboratory but

showed an unsatisfactory pregnancy rate. The swim-up method is the best choice for IUI.

Keywords intrauterine insemination, Puresperm, sperm preparation, swim-up

Intrauterine insemination (IUI) is the most cost-effective treatment for unexplained andmoderate male-factor subfertility [22], achieving pregnancy rates of 10% to 18% per cyclecompared with a spontaneous pregnancy rate of 2% per cycle [19]. In the past, Percoll hadbeen a standard technique for sperm preparation, as it had the best efficiency in selectingmotile sperm with good fertilization ability [7]. However, late in 1996, serious concern wasexpressed about the polyvinylpyrrolidone (PVP) component and endotoxin levels ofPercoll [21]. It was replaced by silica stabilized with covalently bound hydrophilic salinesuch as Puresperm. Today, the swim-up method has become the most widely used tech-nique in most infertility laboratories. Although the sperm recovery rate is lower with theswim-up technique than with density gradient centrifugation, it consistently producesbetter suspensions with higher sperm velocity and greater proportions of sperm with intactacrosomes and normal morphology [17]. In this study, we compared the pregnancy ratesand characteristics of semen prepared by these four methods.

Address correspondence to Gwo-Jang Wu, MD, PhD, Division of Endocrinology and Infertility, Department

of Obstetrics and Gynecology, Tri-Service General Hospital, No. 325, Section 2, Cheng-Gung Road, Neihu,

Taipei 114, Taiwan, R.O.C. E-mail: ghsun@mail.ndmctsgh.edu.tw

Archives of Andrology, 50:139–143, 2004Copyright # Taylor & Francis Inc.ISSN: 0148-5016 print / 1521-0375 onlineDOI: 10.1080/01485010490425566

139

Syst

Bio

l Rep

rod

Med

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

SUN

Y S

tate

Uni

vers

ity o

f N

ew Y

ork

at S

tony

Bro

ok o

n 12

/21/

14Fo

r pe

rson

al u

se o

nly.

MATERIAL AND METHODS

317 cycles of IUI were evaluated during a 5-year period from 1998 to 2002. Womenwho had at least one patent fallopian tube by examination with hysterosalpingography(HSG) but failed to get pregnant with at least three courses of clomiphene citrate treat-ment were recruited, excluding couples who did not have an infertility problem but came forY-chromosome selection. All of these couples received controlled ovarian hyperstimula-tion with human menopausal gonodotropin (with or without recombinant FSH) com-bined with human chorionic gonadotropin. The semen samples were collected after 48 to72 hours of abstinence. Ejaculates were left to liquefy for 20 to 30min. About 50 ml wasloaded in a microcell slide for the computer-assisted semen analysis (CASA) system(Image House, Denmark). Total concentration, motile and immotile sperm percentages,hyperactive sperm percentage, average path velocity (VAP), straight-line velocity (VSL),straightness (STR¼VSL=VAP), and amplitude of lateral head displacement (ALH) weredetermined. A two-layer Percoll solution was prepared as follows: 90% Percoll was madeby diluting 9.0ml of Percoll with 1.0ml of buffer (18% NaCl); 45% Percoll was made bydiluting 4.5ml of Percoll with 0.5ml of buffer and 5ml of human tubal fluid=bovine serumalbumin. The ready-to-use Puresperm (Nidacon, Gteborg, Sweden) solution was preparedwith two-layer gradients in 80% and 55% Puresperm. For the swim-up method, we used amultiple-tube direct method to decrease the potential adverse effects of centrifugation thatmay result in sperm damage and a decreased pregnancy rate [16]. For the albumin method,10% human serum albumin was added to the medium with procedures similar to those ofthe swim-up method. Sigma-plot 8.0 with the Student t-test was used compare the char-acteristics of the semen before and after preparation. For comparison of differences in thecharacteristics of semen prepared by the Percoll, Puresperm, and swim-up methods andthose of the albumin method, PROC GENMOD (SAS 8.2) was performed. Multiplelogistic regressions were used for analyzing pregnancy outcome in relation to semencharacteristics. A P value <0.05 was considered statistically significant.

RESULTS

The albumin and Percoll methods had the highest pregnancy rates: up to 12% per cycle,followed by swim-up (9.6%) and Puresperm (7.4%) (Table 1). With all of these methods,there were statistically significant differences in most of the characteristics (P< 0.05) offresh vs. prepared semen, the exceptions being VSL with each method and total con-centration with Puresperm. The Percoll method produced the best VAP and VSL. ThePuresperm method revealed a statistically significant difference in total concentration and

Table 1. Overall pregnancy outcome

Albumin Percoll Puresperm Swim-up

No of patient 38 64 51 91

No of cycles treated 48 � 1.26 86 � 1.34 68 � 1.33 115 � 1.26

Female age 31 � 7.1 33 � 6.9 32 � 6.2 31 � 8.5

Pregnancy rate (%) 12.5 12.7 7.4 9.6

140 S.-S. Ren et al.

Syst

Bio

l Rep

rod

Med

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

SUN

Y S

tate

Uni

vers

ity o

f N

ew Y

ork

at S

tony

Bro

ok o

n 12

/21/

14Fo

r pe

rson

al u

se o

nly.

hyperactive sperm percentage, while the swim-up method had the best yield of motilesperm. After adjusting for age, method, and semen characteristics by multiple logisticregressions, we found only VSL to be significantly correlated with pregnancy (P¼ 0.0488).With an increase in VSL of 1 mm=s, the probability of pregnancy doubled.

DISCUSSION

The Percoll method produced superior VAP and VSL, which might explain the higherpregnancy rate than was obtained by the other two methods. The same result has beenreported by Jaroudi et al. [10]. However, because preparation with polyvinylpyrrolidine-coated silica particles may be harmful to human gametes, Percoll has been replaced byPuresperm [3, 5]. In our study, two-layer Puresperm did produce a better recovery ratethan Percoll, but motility and other semen characteristics were not superior. Preparationsobtained by two-layer Puresperm showed a significant decrease in spermatozoa motilitycompared with those obtained by Percoll and four-layer Puresperm [2]. Three-layerPuresperm produced a higher percentage of motile spermatozoa and better sperm mor-phology than two-layer Puresperm [4]. These findings may account for the fact that Percollwas associated with a significantly higher pregnancy rate than two-layer Puresperm in ourstudy.

Although the swim-up technique was considered to provide the greatest amount ofmotile and hyperactive sperm with a high percentage of normal spermatozoa in mor-phology, the centrifugation of sperm before swim-up may generate reactive oxygen species(ROS), which can cause lipid peroxidation membrane damage and a four-fold increase insperm DNA damage [1, 13]. Twigg et al. also reported that dramatic decline of spermmotility caused by ROS impairs the fertility ability of sperm [23]. However, Younglaiet al., who used the TUNEL assay, did not find anymore DNA damage after centrifugationfor the swim-up technique [24]. Nevertheless, we used the direct swim-up method withmultiple tubes to increase the recovery rate and avoid the ROS effect. The preparationprovided the best motile sperm percentage, a finding compatible with the resultsdemonstrated by other researchers.

The CASA provides high precision and quantitative data on sperm kinetics. Severalpapers have emphasized the correlation between the pregnancy outcome and semencharacteristics estimated by CASA [8, 12, 15, 18]. VCL and hyperactive sperm percentagecorrelate well with fertilizing ability in vitro [9]. In our study, only VSL had a statisticallysignificant correlation with the pregnancy rate. This same result was first reported by Marcet al. [14], who described the major factor controlling sperm motility as being the avail-ability of ATP produced by mitochondria [6]. When mitochondrial DNA suffers a deletionat the 4977-bp site, sperm motility is diminished [11] and VSL is reduced. Sun et al. alsofound a negative association between DNA fragmentation and the fertilization rate [20].Therefore, it is reasonable that a better VSL will increase the pregnancy rate.

We obtained an acceptable pregnancy rate of 11% with a limited ovarian stimulationprotocol that was only one sixth the cost of IVF at our hospital. The swim-up method isthe best method for IVF, as has been demonstrated by several groups. Because thetwo-layer Puresperm method did not produce a satisfactory pregnancy rate, the swim-upmethod is the choice in our laboratory for IUI, as for IVF.

Sperm Preparation in IUI 141

Syst

Bio

l Rep

rod

Med

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

SUN

Y S

tate

Uni

vers

ity o

f N

ew Y

ork

at S

tony

Bro

ok o

n 12

/21/

14Fo

r pe

rson

al u

se o

nly.

REFERENCES

1. Aitken RJ, Clarkson JS (1988): Significance of reactive oxygen species and antioxidants in

defining the efficacy of sperm preparation techniques. J Androl 9:367–376.2. Brita S, Kersti L (2000): The use of Silane-coated silica particles for density gradient cen-

trifugation in in-vitro fertilization. Hum Reprod 15:857–860.

3. Centola GM, Herko R, et al. (1998): Comparison of sperm separation methods: Effect onrecovery, motility, motion parameters, and hyperactivation. Fertil Steril 70:1173–1175.

4. Chen MJ, Bongso A (1999): Comparative evaluation of two density gradient preparations for

sperm separation for medically assisted conception. Hum Reprod 14:759–764.5. Claassens OE, Menkveld R, Harrison KL (1998): Evaluation of three substitutes for Percoll in

sperm isolation by density gradient centrifugation. Hum Reprod 13:3139–3143.6. Cummins J, Yovich J (1993): Semen motility enhancement in vitro. Semen Reprod Endocrinol

11:56–71.7. Elglert Y, Van DBM, Rodesch C (1992): Comparative auto-controlled study between swim-up

and Percoll preparation of fresh semen samples for in-vitro fertilization. Hum Reprod

7:399–402.8. Fetterolf PM, Rogers BJ (1990): Prediction of human sperm penetrating ability using compu-

terized motion parameters. Mol Reprod Dev 27:326–331.

9. Hirano Y, et al. (2001): Relationships between sperm motility characteristics assessed by thecomputer-aided sperm analysis (CASA) and fertilization rates in vitro. J Assist Reprod Genet18:213–218.

10. Jaroudi KA, et al. (1993): Percoll semen preparation enhances human oocyte fertilization in

male-factor infertility as shown by a randomized cross-over study. Hum Reprod 9:1438–1442.11. Kao S, Chao H, Wei H (1995): Mitochondrial deoxyribonucleic acid 4977-bp deletion is asso-

ciated with diminished fertility and motility of human sperm. Biol Reprod 52:729–736.

12. Liu DY, Clarke GN, Baker HGW (1988): Relationship between motility assessed with theHamilton-Thorn Motility Analyzer and fertilization rates in vitro. J Androl 12:231–239.

13. Lopes S, et al. (1998): Reactive oxygen species: potential cause for DNA fragmentation in

human spermatozoa. Hum Reprod 13:896–900.14. Marc VB, et al. (1998): A first prospective study of the individual straight line velocity of the

spermatozoon and its influences on the fertilization rate after intracytoplasmic sperm injection.

Hum Reprod 13:3103–3107.15. Mathur S, et al. (1986): A computerized sperm motion analysis. Fertil Steril 46:484–488.16. Mortimer D (1991): Sperm preparation techniques and iatrogenic failures of in-vitro fertiliza-

tion. Hum Reprod 6:173–176.

17. Oehninger S, et al. (1990): Relationship between morphology and motion characteristics ofhuman spermatozoa in semen and in the swim-up sperm fractions. J Androl 11:446–452.

18. Parinaud J, et al. (1996): Are the characteristics of spermatozoa in the insemination medium

useful for predicting in-vitro fertilization results? Int J Androl 19:103–108.19. Stone BA, et al. (1999): Determinants of outcome of intrauterine insemination: analysis of

outcomes of 9963 consecutive cycles. Am J Obstet Gynecol 180:1522–1534.

20. Sun J, Juriscova A, Casper R (1997): Detection of deoxyribonucleic acid fragmentation inhuman sperm: correlation with fertilization in vitro. Biol Reprod 56:602–607.

21. Svalander PC, Lundin K, Holmes PV (1995): Endotoxin levels in Percoll density-gradient mediaused to prepare sperm for human IVF treatment. Hum Reprod 10 (abstract book 2):130.

22. Tariq M, Michael C (2002): The use of intrauterine insemination in Australia andNew Zealand. Hum Reprod 17:956–959.

142 S.-S. Ren et al.

Syst

Bio

l Rep

rod

Med

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

SUN

Y S

tate

Uni

vers

ity o

f N

ew Y

ork

at S

tony

Bro

ok o

n 12

/21/

14Fo

r pe

rson

al u

se o

nly.

23. Twigg J, Fulton N, Gomez E (1998): Analysis of the impact of intracellular reactive oxygen

species generation on the structural and functional integrity of human spermatozoa: Lipid per-oxidation, DNA fragmentation and effectiveness of antioxidants. Hum Reprod 13:1429–1436.

24. Younglai EV, et al. (2001): Sperm swim-up techniques and DNA fragmentation. Hum Reprod

16:1950–1953.

Sperm Preparation in IUI 143

Syst

Bio

l Rep

rod

Med

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

SUN

Y S

tate

Uni

vers

ity o

f N

ew Y

ork

at S

tony

Bro

ok o

n 12

/21/

14Fo

r pe

rson

al u

se o

nly.