Microsurgical fertilization procedures: The absence of stringent criteria for patient selection

Journal of Assisted Reproduction and Genetics, Vol. 9, No. 3, 1992 CLINICAL ASSISTED REPRODUCTION

Microsurgical Fertilization Procedures: The Absence of Stringent Criteria for Patient Selection

JACQUES COHEN, 1'2 MINA ALIKANI, l ALEXIS ADLER, 1 ALAN BERKELEY, 1 0 W E N DAVIS, x TONI A. FERRARA, 1 MARGARET GRAF, 1 JAMIE GRIFO, 1 HUNG-CHING LIU, a HENRY E. MALTER, t ADRIENNE M. REING, 1 MICHAEL SUZMAN, 1 BETH E. TALANSKY, 1 JANET TROWBRIDGE, a and ZEV ROSENWAKS 1

Submitted: September 4, 1991 Accepted: Match 6, 1992

Subzonal sperm insertion and partial zona dissection were applied in 250 in vitro fertilization cycles in couples (n = 200) with abnormal semen analyses; 61 clinical pregnancies were established (24% per egg retrieval). Pa- tients were selected without using minimal cutoff criteria. The study included patients with 0% normal sperm forms (strict criteria), no motile sperm (but some live cells), and sperm counts which could be assessed only after centrifugation. Patients were categorized into three subsets. Group A (n = 116 cycles)failed to fertilize in a previous cycle. Group B (n = 40) was excluded from IVF due to the severity o f sperm profiles, such as a maximum o f 2% normal forms. Group C (n = 94) constitutes those patients for whom a standard cycle could possibly result in failure. Monospermic fertilization rates were 18% (A), 19% (B), and 24% (C). The incidences o f embryo replacement were 63% (A), 53% (B), and 69% (C). Rates o f clinical pregnancy were 22% (A), 23% (B), and 28% (C). The presence of one, two, or three semen abnormalities did not correlate with the outcome o f microsurgical fertilization. Twenty-two percent o f patients with combined oligoasthenoteratozoospermia became pregnant. More- over, ongoing pregnancies were established in instances with 0% normal sperm forms and no progressively motile spermatozoa. It is concluded that stringent cutoff criteria may not be necessary when both partial zona dissection and subzonal sperm insertion are performed efficiently.

i Center for Reproductive Medicine and Infertility, Department of Obstetrics and Gynecology, New York Hospital-Cornell University Medical Center, New York, New York 10021.

2 To whom correspondence should be addressed at the Gamete and Embryo Research Laboratory, Cornell University Medical College, P.O. Box 30, 1300 York Avenue, New York, New York 10021.

KEY WORDS: subzonal sperm insertion; partial zona dissection; sperm factors; oligoasthenoteratozoospermia.

INTRODUCTION

Progress in the application of assisted reproductive technology for treatment of male-factor infertility is dependent largely on the efficiency of new microsurgical fertilization procedures. Thus far, two such methods have been assessed in small series of in- vestigations (1-6). The first involves the creation of an artificial gap in the zona pellucida (ZP) through which spermatozoa may penetrate following insemination (zona drilling, partial zona dissection; PZD) (2,7). In a second procedure, oocytes are not inseminated, but a single or a few spermatozoa are inserted into the perivitelline space (subzonal sperm insertion, SZI) (1,8). There is potentially a large patient population which may benefit from these methods (9). Microsurgical fertilization techniques may be applied in couples in which the male partners have few spermatozoa that are able to fuse with or penetrate the ZP in vitro.

We have now applied PZD and SZI with success in over 250 in vitro fertilization (IVF) cycles for patients with consistent abnormal semen analyses. Though it has generally been the aim to perform the three insemination techniques (IVF, PZD, SZI) in parallel using sibling oocytes, this strategy was frequently hindered by the limited number of oocytes or spermatozoa (9). Moreover, many patients had undergone IVF previously and failed to fertilize zona-intact oocytes. Nevertheless, in most of the patients, two or three of the insemination techniques have been applied. It was found, with few

197 1058-0468/92/0600-0197506.50/0 © I992 Plenum Publishing Corporation

198 COHEN ET AL.

exceptions, that microsurgical fertilization yielded higher rates of monospermic fertilization than regular insemination (2,6,10-12). Moreover, fertilization following SZI was significantly higher than with PZD in patients who had few spermatozoa available for insemination (6,13). However, spermatozoa from another, less well-defined group of patients seemed to be successful only following PZD (6). It was shown that the effects of both methods of microsurgical fertilization were also additive in terms of implantation. PZD embryos preferen- tially implanted in women whose partners had mod- erate teratozoospermia, whereas SZI embryos implanted more frequently in instances of severe teratozoospermia (6).

One of the potential advantages of SZI is that it allows for assisted fertilization to be performed when only a few spermatozoa can be retrieved (1). Consequently, potential patients could include those with complete absence of motility and extreme oligozoospermia (<1 x 106/ml) and others with abnormally shaped spermatozoa. This policy of not employing cutoff criteria has been in effect in this program for over a year. The aim of this study was therefore to investigate the prognoses of patients with severely abnormal semen analyses and to determine whether low cutoff limits should be implemented for the application of assisted fertilization. The results of the micromanipulation procedures in the first 250 cycles were correlated with fertilization and implantation using a variety of analyses. For the first breakdown, absolute figures for motility, sperm concentration, and number of normally shaped spermatozoa were correlated with the outcome of assisted fertilization. In a second analysis, patients with a single semen abnormality (either oligozoospermia, asthenozoospermia, or teratozoospermia) were compared to those with two or three abnormalities.

MATERIALS AND METHODS

Patient Grouping and Semen Criteria

This report includes 250 microsurgical fertilization cycles (200 patients) performed between Octo- ber 1989 and July 1991. Approval for clinical application of both techniques was granted by the New York Hospital-Cornell Medical Center Committee of Human Resources. PZD and SZI were indicated for these patients because semen analyses were

consistently abnormal. With the exception of seven exclusively oligozoospermic (and/or asthenozoospermic) cases, all patients had teratozoospermia (~<10% normal sperm forms according to Kruger's strict criteria) (14). The other criteria for consideration as a patient with abnormal semen analysis have been published previously (~<10 x 106/ml for oligozoospermia and ~<20% motile spermatozoa or the absence of progression for asthenozoospermia) (6,9). There is no lower limit for accep- tance into the microsurgical fertilization program for teratozoospermia (patients with 0% normal forms are included) and asthenozoospermia (0% motility is acceptable provided that some spermatozoa are alive when stained with eosin). The lower limit for oligozoospermia cannot be accurately determined, as sperm counts were not measurable in a number of patients and spermatozoa were isolated only after centrifugation. In addition to these criteria, we have defined four groups of patients who may be treated with microsurgical fertilization. One of these groups is not studied in the present inves- tigation and includes idiopathic patients with appar- ently normal semen analyses, who failed to fertilize zona-intact oocytes during regular IVF cycles. The results of the first 30 such cases have been presented elsewhere (15). The remaining patients with consistently abnormal semen analyses have been defined previously and can be subdivided into the following three groups (6).

The first group (Group A; n = 116 cycles) had abnormal semen analysis (0-10% normal sperm forms according to Kruger's strict criteria, 0-20% motility and/or 0.1-10 × 10 6 spermatozoa/ml) and failed to fertilize all oocytes in previous IVF cycles.

The second group of patients (Group B; n = 40) had not been accepted for regular IVF by other programs, including our own. These patients' semen analyses were considered highly abnormal: 0-2% normal sperm forms (Kruger's strict criteria) in combination with either extreme oligozoospermia (~<0.1-5 x 106/ml) or extreme asthenozoospermia (0-10% motile). The upper limit of 2% normal forms was chosen since such sperm profiles were not compatible with viable pregnancy in regular IVF cycles. Such patients were advised to have microsurgical fertilization rather than a standard IVF cycle, although zona-intact oocytes were occasionally included when sufficient numbers of eggs were available.

The third group (Group C) of 94 cycles had not attempted IVF previously (some were not accept-

Journal o f Assisted Reproduction and Genetics, Vol. 9, No. 3, 1992

SEMEN FACTORS AND MICROSURGICAL FERTILIZATION 199

able to other IVF programs) and all had male-factor infertility of intermediate severity. These men had teratozoospermia (3-10% normal sperm forms) and their semen was considered oligozoospermic (0.5- 20 x 106/ml) and/or asthenozoospermic (1-20% motile with reduced linear progression). Since the chances of fertilization could be somewhat reduced among these patients, micromanipulation was performed on some of their oocytes, while other oocytes were left intact. Many other male-factor IVF patients with similar profiles were not treated with assisted fertilization. The prepared sperm sus- pensions were considered suitable for standard IVF if sufficient numbers were retrieved and forward progression was considered satisfactory. More- over, some male-factor couples did not consent to micromanipulation.

Follicular stimulation has been described elsewhere and was performed using either a sequential clomiphene citrate/human menopausal gonadotropin protocol or gonadotropin stimulation following down-regulation with leuprolide acetate (16,17).

Sperm Preparation and Embryo Culture

Semen was prepared according to sedimentation, multitube layover, or Percoll techniques (6,18). Oocyte and embryo culture methodology, including the use of the mineral oil microdrop method, has also been described in detail (19). Human tubal fluid (HTF) culture medium supplemented with 10% maternal serum was usually used, but a small series of procedures was performed with supplemented Ear- le's solution (20). Serum supplementation was not used during fertilization if the female partner had antibodies against spermatozoa or when normal- colored serum could not be isolated. In these instances, 6% human serum albumin (HSA; Plas- manate, Cutter Biological, Miles Inc., IN) was used for the first 24 hr of the procedure. Four to seven hours postretrieval, oocytes were exposed to 0.1% hyaluronidase (sheep testes Type III, Sigma, St. Louis, MO) in modified phosphate-buffered saline (PBS, with pyruvate and penicillin) supplemented with 6% HSA. Excess corona cells were usually removed with a narrow-bore glass pipette. Oocytes were washed four times in culture medium and gently manipulated with a probe in order to visual- ize and select those with a clear first polar body.

All micromanipulations were performed in pre- warmed depression slides in 2- to 3-fxl droplets con- sisting of modified PBS supplemented with 0.1 M

sucrose and 6% HSA (19). The droplets were contained under warm mineral oil. A maximum of three eggs was pipetted into a depression slide. Sucrose was removed by washing the eggs four times following micromanipulation.

Micromanipulation

The actual procedure for partial zona dissection has been described in detail (6,19). During this study, hydraulic Narishige micromanipulators (Na- rishige, Tokyo) were used in conjunction with an inverted diaphot Nikon microscope using Normaski optics (Nikon, Tokyo). Holding pipettes and needles were pulled on a horizontal Sachs Flaming mi- cropipette puller (Sutter Instrument Co., Novato, CA). Beveled sperm insertion pipettes made from glass capillary tubing (Sutter Instrument Co., Type B 10075-10) were pulled, broken on the microforge, beveled on a Narishige beveler, and washed with hydrofluoric acid and sterile water. An extra sharp point was made on the lowest part of the bevel by drawing out the tip using the microforge.

Sperm insertion needles were front-loaded with spermatozoa by aspiration from the edges of 2- to 4-~1 oil-covered droplets contained in a depression slide. Sperm insertion needles were tangentially inserted into the upper region (" 12 o'clock" position) of the perivitelline space, with the holding pipette at "9 o'clock." The number of inserted spermatozoa ranged from one to five. In the absence of motile spermatozoa, more spermatozoa were inserted. The number of spermatozoa used was independent of the degree of progression or the severity of teratozoospermia. The amount of fluid inserted was kept to a minimum. Oocytes treated with subzonal sperm insertion were incubated in sperm-free droplets separated from inseminated oocytes in which PZD had been performed.

RESULTS

Overall Micromanipulation Results

A total of 2199 mature eggs was obtained during 250 IVF cycles. Of these, 21% were fertilized. The highest rate of monospermic fertilization (24%; 236/ 970) occurred in Group C patients, and the lowest rate (18%, 165/908) in Group A patients (Fig. 1). Though the rates between these groups, as well as between B and C, were significantly different (chi-

Journal of Assisted Reproduction and Genetics, Vol. 9, No. 3, 1992

200 COHEN ET AL.

65194

159/250 73/116

236/970

21/40 Iiiiiil li]il liil

li!ill lii iil lii l liiii!l

lii il

26194

26/116 9/40 --J'-~--61/250 11/46 l 32/155 ,...;"~. 39/180 82/381

_ , - _ _ I M

~;'.. ...;.,.

, ~ , ~ - - - -

U L~.t/, ,¢~

monospermic fertilization/egg replacements per cycle

previous fertilization ] unnacceptable for [ ] Group A failure with regular IVF I [ ] Group B regular IVF

(n=116) (n=40)

Fig. 1. Overall results of the application of microsurgical fertilization Statistical comparisons are

clinical pregnancy per cyc le implantation (FHB/embryo)

reduced prognosis I [ ] Group C male factor [ ] Total (n=250)

(n=94)

in 250 in vitro fertilization cycles with abnormal semen profiles. presented in the text.

square analysis; P < 0.01), fertilization in all groups selected for micromanipulation was considerably reduced from that of other IVF patients in this program. The incidence of polyspermy in PZD and SZI zygotes was nearly identical. Five percent of PZD eggs (35/718) and 4% of SZI eggs (37/899) were polyspermic. The incidence of replacement (the frequency of having at least one embryo for transfer) was 66% (159/250) for all microsurgical cycles combined. The lowest rate of 53% was found in Group B patients, whereas 63% of patients who failed to fertilize zona-intact eggs in previous IVF cycles were able to have a replacement. Significantly more replacements occurred in Group C than in Group B patients (P < 0.05). The incidences of clinical pregnancy per egg retrieval (defined by the presence of a fetal heartbeat) of all three groups were in excess of 20%, despite the relatively low rates of fertiliza-

tion. Sixty-one couples (22%) had a clinical pregnancy, and eight of these miscarried. At this point, 36 patients have delivered 51 healthy babies. Preg- nancies of 17 patients are still ongoing. Of the 463 monospermically fertilized eggs, some were cryopreserved, whereas others either did not divide or underwent abnormal development. A total of 381 embryos was replaced in 159 patients. The rates of embryonic implantation did not differ among the three groups and all were in excess of 20% (82 fetal heartbeats). Interestingly, 11 of 46 (24%) embryos in Group B implanted despite the upper cutoff limit of 2% normal sperm forms.

The majority of pregnancies (37/61; 60%) was established when only microsurgically fertilized embryos were replaced (Fig. 2). Four pregnancies were obtained when zona-intact control embryos were replaced. The remainder of the pregnancies

Journal o f Ass is ted Reproduct ion and Genetics, Vol. 9, No. 3, 1992


GROUP A

]PZD ONLY (n=7) ] [PZD/SZUZP-INTACT (n=3) I

(n=l)]

'2

[SZI ONLY (n=l 5) ']

GROUP B i

I SZI ONLY (n=4) J

G ]PZD/SZI/ZP-INTACT (n=S) I

I GROUP C [

~ D ONLY (n=5) ]

I (n=l) 1

- /

/ t / / j l , \ - ~ . ~ [SZI ONLY (n=l)J

]PZD/SZI/ZP-INTACT (n=19) ]

ISZI ONLY (n=20)~

PZD/S~ZI (n=5) ]

TOTAL i

]PZD ONLY (n= 12) I

A [PZD/SZIYZP-INTACT (n=24) ]

Fig. 2. Contribution of the three insemination procedures to the pregnancies in the three groups of patients. Group A represents patients in whom standard IVF has failed before. Group B patients are not generally acceptable for IVF and Group C patients represent those first-time IVF patients with an abnormal semen analysis in whom the sperm preparation is considered very poor. PZD, partial zona dissection; SZI, subzonal insertion; ZP, zona pellucida.

were derived from replacements that included both microsurgical and unmanipulated embryos. The vast majority (23/26; 81%) of pregnancies in Group A was from microsurgically fertilized embryos; 7 from PZD, 15 from SZI and 1 from a combination of both procedures. The fact that four pregnancies included zona-intact embryos indicates that previous failure of fertilization does not preclude fertilization in a subsequent IVF cycle. The nine pregnancies in Group B were derived from replacements which al- ways included SZI embryos. Based on previous ob- servations, it can be argued that the PZD and ZP- intact embryos in this group did not implant (6,14). The vast majority of pregnancies including ZP- intact embryos was, not surprisingly, found in


Group C patients (Fig. 2). Though one could postu- late that micromanipulation should not be initially applied in these patients, seven of the patients became pregnant from uterine placement of micromanipulated embryos only.

The ability of microsurgically fertilized embryos to implant was analyzed by excluding replacements which included regular IVF embryos (Table 1). Fol- lowing 111 replacement cycles, 44 (40%) of the patients had a positive beta-human chorionic gonadotropin ([3-hCG) and 37 (33%) had a clinical pregnancy. Embryonic implantation ranged from 18 to 23%, indicating that the viability of embryos is not impaired by micromanipulation. The lowest rate of 18% was found in the PZD group. This was due to

202 COHEN ET AL.

Table I. Implantation in Cycles in Which Only Micromanipu- lated Embryos Were Replaced

Incidence of Incidence of Mode of pregnancy implantation

manipulation per replacement per embryo

PZD embryos only replaced 12/44 (27%) 14/77 (18%)

SZI embryos only replaced 20/57 (35%) 25/111 (23%)

Both PZD and SZI embryos replaced 5/10 (50%) 5/26 (19%)

the inability of most PZD embryos from extreme teratozoospermic patients to implant, as has been described previously (6).

Sperm Count and Outcome of Microsurgical Fertilization

Results of the microsurgical procedures as a function of the sperm count are presented in Table II. Significant trends could not be demonstrated since the lowest rates of fertilization (18-20%) and implantation (16-21%) were found in the lowest range (~<l x 106/ml) and the highest range (>20 x 106/ml) of sperm concentrations. Consequently, the sperm concentration or the severity of oligozoospermia cannot be considered a prognostic factor for the outcome of microsurgically assisted fertilization. Though a relatively large group of cycles (n = 99) appeared to have been performed with spermatozoa from men who were not oligozoospermic, all of them were teratozoospermic or asthenozoospermic. The frequency of oligozoospermia presented here is probably reduced due to the use of split ejaculates in many instances (semen counts presented here are determined on the day of egg retrieval).

Motility and Outcome of Microsurgical Fertilization

Results of the microsurgical procedures as a function of the percentage motility determined in the

semen sample obtained on the day of egg collection are presented in Table Ill. Again, the frequency of asthenozoospermia is probably reduced due to the use of split ejaculates in many instances. This may explain the fairly large group of ejaculates in which more than 30% of the spermatozoa were motile. However, all these samples were considered teratozoospermic. The rate of monospermic fertilization seems to be unaffected by the proportion of motile spermatozoa in the original semen sample. Embryos from patients with extreme forms of asthenozoospermia (~<5% motility) appeared to have a lower ability to implant (2/18; 11%). However, it is likely that this is due to the sample size of this group, since embryos from other patients with extreme asthenozoospermia (6-10% motile) implanted frequently (10/23; 43%). Consequently, it may be concluded that the severity of asthenozoospermia is not predictive of the outcome of assisted fertilization. Indeed, microsurgical fertilization has been accomplished in two patients with complete absence of motility. In another azoospermic patient who underwent epididymal aspiration, all spermatozoa lacked forward progression. A single embryo was obtained following SZI of 16 eggs and the patient has delivered a healthy child.

Percentage Normal Sperm Forms and Outcome of Microsurgical Fertilization

Results of the microsurgical procedures as a function of the percentage normal sperm forms (on the day of egg collection) are presented in Table IV. Significant correlations could not be established between percentage normal sperm forms and fertilization: Pregnancy and embryonic implantation results were not found to be correlated with teratozoospermia either. Consequently, the severity of teratozoospermia cannot be used as a prognostic factor for predicting the outcome of microsurgically assisted fertilization. It is likely, that this is due to the efficiency of SZI in extreme teratozoospermic

Table II. Results of Microsurgical Fertilization Expressed as a Function of the Sperm Concentration of the Semen Sample

Sperm concentration No. of Monospermic Incidence of Incidence of Fetal heartbeat (x 106/ml) cycles fertilization replacements pregnancy per embryo

~<1 38 70/387 (18%) 20/38 (53%) 6/38 (16%) 11/52 (21%) 2-5 28 50/235 (21%) 15/28 (54%) 9/28 (32%) 10/37 (27%) 6-10 43 85/362 (23%) 33/43 (77%) 14/43 (33%) 18/76 (24%)

11-20 42 85/320 (27%) 28/42 (67%) 14/42 (33%) 17/66 (26%) >20 99 179/903 (20%) 64/99 (64%) 18/99 (18%) 25/154 (16%)



Table III. Results of Microsurgical Fertilization Expressed as a Function of Sperm Motility in the Semen Sample

Percentage No. of Monospermic Incidence of Incidence of Fetal heartbeat motility cycles fertilization replacements pregnancy per embryo

~<5 16 29/124 (23%) 9/16 (56%) 2/16 (13%) 2/18 (11%) 6-10 30 23/261 (9%) 13/30 (43%) 6/30 (20%) 10/23 (43%)

11-20 35 57/310 (18%) 25/35 (71%) 10/35 (29%) 13/54 (24%) 21-30 33 89/335 (27%) 23/33 (70%) 7/33 (21%) 13/64 (20%) >30 137 255/1089 (23%) 84/127 (66%) 34/137 (25%) 41/213 (19%)

cases. Only 7% of PZD eggs were fertilized in patients with 0% normal sperm forms, whereas 20% of eggs were fertilized in such patients when SZI was performed. Moreover, it was previously shown that the ability of SZI embryos to implant from patients with extreme te ra tozoospermia (0-2% normal forms) was not impaired, whereas PZD embryos from similar patients rarely implanted (6). Indeed, of the 64 cycles thus far attempted in such patients, 14 (22%) have had a clinical pregnancy.

The Number of Semen Abnormalities and Outcome of Microsurgical Fertilization

Results of microsurgical fertilization expressed as a function of the number of semen abnormalities (ol igozoospermia, as thenozoospermia , teratozoospermia) are presented in Table V. It is obvious from these results that the outcome of microsurgical fertilization cannot be predicted on the basis of the WHO criteria. The presence of one, two, or three abnormalities does not affect the rates of fertilization and pregnancy. This is also demonstrated by examining the data for pregnant patients who had extremely abnormal semen profiles and who were considered unsuitable for IVF. Results of 10 such cases are illustrated in Table VI.

The three insemination options (IVF, PZD, and SZI) are potentially applicable in sibling oocytes of all microsurgical patients. Similarly, the three alternatives were contemplated in these l0 couples.

However, the decision to perform these modes of insemination in each case was based on three vari- ables: (i) the number of oocytes, (ii) the number of spermatozoa, and (iii) the fertilization outcome in previous cycles. In patient 114 for instance, regular IVF had failed twice and PZD had also been unsuc- cessful. Consequently, it was decided to perform SZI on all oocytes. In patient 269, on the other hand, IVF was never attempted due to the absence of normally shaped spermatozoa. However 19 mature eggs were retrieved. This was considered sufficient to perform all three insemination techniques in parallel, though more eggs were allocated to SZI than to the two other procedures, due to the increased probability of implantation.

DISCUSSION

The present findings clearly demonstrate the lack of a correlation between quantitative sperm param- eters and the outcome of assisted fertilization procedures in couples with a history of male-factor infertility. A substantial number of viable pregnancies were established in couples where the male partner's semen analysis fell well below the normal cutoff values for regular assisted conception procedures. Fertilization and pregnancy occurred following the use of spermatozoa without progressive motility and without normal morphology. In some patients, sperm counts were correspondingly re-

Table IV. Results of Microsurgical Fertilization Expressed as a Function of the Percentage of Normal Sperm Forms in the Semen Sample

Percentage normal No. of Monospermic Incidence of Incidence of Fetal heartbeat sperm forms cycles fertilization replacements pregnancy per embryo

0 16 18/123 (15%) 11/16 (69%) 4/16 (25%) 4/18 (22%) 1 20 30/149 (20%) 9/20 (45%) 4/20 (20%) 5/24 (21%) 2 28 28/161 (17%) 18/28 (64%) 6/28 (21%) 8/42 (19%) 3 26 27/134 (20%) 19/26 (73%) 10/26 (38%) 15/49 (31%)

4-5 66 131/543 (24%) 45/66 (68%) 15/66 (23%) 21/96 (22%) 6-8 50 108/472 (23%) 34/50 (68%) 15/50 (30%) 19/88 (22%) >8 35 63/353 (18%) 19/35 (54%) 5/35 (14%) 7/55 (13%)


204 COHEN ET AL.

Table V. Results of Microsurgical Fertilization Expressed as a Function of the Number of Semen Abnormalities a

No. of semen No. of Monospermic Incidence of Incidence of Fetal heartbeat abnormalities cycles fertilization replacements pregnancy per embryo

1 80 144/660 (22%) 47/80 (59%) 17/80 (21%) 21/120 (18%) 2 115 236/1059 (22%) 85/115 (74%) 33/115 (29%) 43/198 (22%) 3 (OAT) 46 69/400 (17%) 24/46 (52%) 10/46 (22%) 15/56 (27%)

a O, oligozoospermia; A, asthenozoospermia; T, teratozoospermia.

duced and spermatozoa could be visualized only after centrifugation. The current results provide ev- idence that spermatozoa from extremely oligoas- thenoteratozoospermic men can produce normal offspring after the application of micromanipulation techniques, even when fertilization previously failed following standard IVF. We were not able to identify, on the basis of conventional semen evalu- ations, subgroups of patients who would not benefit from microsurgical techniques. Though spermatozoa from approximately one-third of the patients failed to fertilize altogether, this could not be correlated with semen analysis.

It seems likely that the current findings are the results of combining PZD and SZI. Previously, it was shown that the techniques complement each other (6,9). Some patients' spermatozoa will fertilize with only one procedure, whereas in other patients in whom both techniques yield embryos, implantation will more likely occur with embryos from one particular micromanipulation procedure. Con- sequently, exclusion of one of the two techniques would more likely result in a correlation with semen factors (5,6,12). A more in-depth analysis of the current cycles demonstrates, for instance, that monospermic fertilization occurred in 7% of the PZD oocytes in instances with 0% normal sperm forms. In similar patients, 20% of eggs fertilized

when SZI was applied. The reverse situation was found in patients with more than 6% normal sperm forms, indicating that, in general, the effect of teratozoospermia on the outcome of assisted fertilization will be negated when both micromanipulation techniques are available.

Though the use of assisted fertilization appears valid when standard IVF has failed (Group A) or when too few spermatozoa are available from insemination (Group B), the issue of patient selection remains unclear (6-6). In addition, the situation may be more complex because some techniques, such as PZD, involve insemination procedures. Dif- ferences in culture systems between laboratories (for instance, microdroplet versus tube method) may then complicate any conclusion regarding the usefulness of the technique (13). Moreover, some laboratories may combine two techniques with the same oocyte (for instance, insemination of SZI eggs), which invalidates any argument (21). In general, we have adopted the following policies con- cerning the treatment of male-factor patients in the microsurgical fertilization program: Oocytes from Group A patients will all be micromanipulated if standard IVF has failed twice before. An internal control with zona-intact oocytes in Group A will be considered when IVF has failed only once and when enough spermatozoa are recovered for insem-

Table VI. Examples of Pregnancies Following Replacement of Microsurgical Embryos in Patients Whose Partners Had Severely Poor Semen Profiles

Semen analysis Fertilization Outcome

Cycle Maternal % % ZP No. of (no. of fetuses no. Group age Count motility normal intact PZD SZI embryos or babies)

85 C 30 24 0 5 0/2 - - 1/16 1 Delivered (x 1) 114 A 40 1 40 2 - - - - 3/12 3 Delivered (x2) 116 A 34 0.3 10 8 0/4 - - 4/11 4 Delivered (x l ) 171 A 30 0.8 25 4 - - - - 4/14 3 Delivered (x2) 186 B 35 2 50 0 0/3 - - 1/3 1 Miscarried (x 1) 187 A 29 0.1 10 0 - - 0/3 1/6 1 Delivered (x 1) 269 B 29 8 35 0 0/4 0/5 4/10 4 Ongoing (x 1) 284 B 37 2 1 7 - - 3/4 3/7 2 Ongoing (x 1) 291 A 37 8 60 0 - - 1/5 2/9 2 Ongoing (x l ) 292 C 33 15 1 5 - - 1/1 2/2 2 Ongoing (x 1)



ination in several microdroplets. Sperm recovery is usually very low in Group B patients. Hence most oocytes are allocated to micromanipulation. The application of micromanipulation in other male- factor cases is based on subjective criteria. Most of them are treated with conventional IVF when progression and numbers are relatively satisfactory after semen preparation. In the remainder (Group C), micromanipulation is performed on only some of the oocytes. The current results demonstrate that selection criteria regarding assisted fertilization procedures should not be absolute. Implantation of zona pellucida-intact embryos in these studies def- initely occurred in a few Group A and in most Group C patients. However, it may also be argued that there would have been fewer pregnancies in Group C patients if micromanipulation had not been performed.

In summary, two main issues must be considered when implementing microsurgical fertilization techniques. First, although results are encouraging, these procedures should be further improved in order to generate better rates of fertilization. How- ever, increased fertilization rates do not necessarily translate into increased embryonic implantation. Thus, it has become apparent from the results of several IVF programs that although fertilization is frequently achieved by methods such as PZD and SZI, pregnancy rates remain low (3,4,11,21,22). This discrepancy in implantation results among different programs has been attributed to differences in patient selection criteria. However, the current study demonstrates that SZI as well as PZD embryos implant frequently, regardless of the semen profile. It is therefore more likely that methodolog- ical dissimilarities among programs account for differences in implantation results.

ACKNOWLEDGMENTS

Elena Kissin, Cindy Anderson, Miriam Feli- ciano, Myriam Jackson, Donna Ludwig, Maureen Hunt, Kim Prey, Mary Hartpence, J. Michael Bed- ford, and Peggy King are gratefully acknowledged for their support of this study.

REFERENCES

1. Laws-King A, Trounson A, Sathananthan H, Kola I: Fertil- ization of human oocytes by microinjection of a single sper-

matozoon under the zona pellucida. Fertil Steril 1987;48: 637-642

2. Cohen J, Malter H, Wright G, Kort H, Massey J, Mitchell D: Partial zona dissection of human oocytes when failure of zona pellucida penetration is anticipated. Hum Reprod 1989; 4:435-442

3. Ng SC, Bongso A, Sathananthan H, Ratnam SS: Microma- nipulation: Its relevance to human in vitro fertilization. Fer- til Steril 1990;53:203-219

4. Fishel S, Jackson P, Antinori S, Johnson J, Grossi S, Versaci C: Subzonal insemination for the alleviation of infertility. Fertil Steril 1990;54:828-835

5. Tucker MJ, Bishop FM, Cohen J, Wiker SR, Wright G: Rou- tine application of partial zona dissection for male factor infertility. Hum Reprod 1991;5:676-681

6. Cohen J, Alikani M, Malter HE, Adler A, Talansky BE, Rosenwaks Z: Partial zona dissection or subzonal insertion: microsurgical fertilization alternatives based on evaluation of sperm and embryo morphology. Fertil Steril 1991;56:696- 706

7. Gordon JW, Talansky BE: Assisted fertilization by zona drilling: A mouse model for correction of oligospermia. J Exp Zool 1986;239:347-354

8. Lassalle B, Coutot AM, Testart J: In vitro fertilization of hamster and human oocytes by microinjection of human sperm. Gamete Res 1987;16:6%78

9. Cohen J: A review of clinical microsurgical fertilization. In Micromanipulation of Human Gametes and Embryos, J Co- hen, HE Malter, J Grifo, B Talansky (eds). New York, Raven Press, 1991, pp 163-190

10. Cohen J, Malter H, Talansky B, Tucker M, Wright G: Ga- mete and embryo micromanipulation. In The Impact of IVF on the Infertile Couple, RP Marrs, L Speroff (eds). New York, Thieme, 1990, p 290

11. Garrisi GJ, Talansky BE, Brunfeld L, Sapira V, Navot D, Gordon JW: Clinical evaluation of three approaches to micromanipulation-assisted fertilization. Fertil Steril 1990;54: 671-677

12. Cohen J, Talansky BE, Malter HM, Alikani M, Adler A, Reing A, Berkeley A, Graf M, Davis O, Liu H, Bedford JM, Rosenwaks Z: Microsurgical fertilization and teratozoospermia. Hum Reprod 1991;6:118-123

13. Sakkas D, Diotallevi L, Ferraretti A, Gianaroli L: Prelimi- nary results on the comparison of sperm microinjection and partial zona dissection as a treatment for male factor infertility. 7th World Congress on IVF and Assisted Procreation, 119, abst 107, p 144

14. Kruger TF, Acosta AA, Simmons KF, Swanson ILl, Matta JF, Oehninger S: Predictive value of abnormal sperm morphology in in vitro fertilization. Fertil Steril 1988;49:11-121

15. Cohen J, Malter H, Adler A, Alikani M, Berkeley A, Davis O, Graf M, Reing A, Talansky B, Rosenwaks Z: Reduced implanting ability of microsurgical embryos from couples with normal semen analysis who failed to fertilize following standard IVF. 7th World Congress on IVF and Assisted Pro- creation, 1991, Abstr 70, p 132

16. Cohen J, DeVane GW, Eisner CW, Kort HI, Massey JB, Norbury SE: Cryopreserved zygotes and embryos and en- docrinologic factors in the replacement cycle. Fertil Steril 1988;50:61-67

17. Droesch K, Muasher SJ, Brzyski RG, Jones GS, Simonetti


206 COHEN ET AL.

S, Liu H-C, Rosenwaks Z: Value of suppression with a gonadotropin-releasing hormone agonist prior to gonadotropin stimulation for in vitro fertilization. Fertil Steril 1989;51:292

18. Ord T, Patrizio P, Marello E, Balmaceda JP, Asch RH: A new method of semen preparation for IVF in severe male factor infertility. Hum Reprod 1990;5:987-989

19. Cohen J: A practical guide to microsurgical fertilization in humans. In Micromanipulation of Human Gametes and Em- bryos, J Cohen, HE Malter, J Grifo, B Talansky (eds). New York, Raven Press, 1991; pp 140-162

20. Quinn P, Warnes GM, Kerin JF, Kirby C: Culture factors

affecting the success rate of IVF and embryo transfer. Ann NY Acad Sci 1985;442:195-204

21. Fuscaldo G, Sobieszczuk D, Trounson AO: Improved fertilization rates following microinjection of human spermatozoa pretreated with 2'-deoxyadenosine and pentoxfylline. 7th World Congress on IVF and Assisted Procreation, 1991, Abstr 110, p 145

22. Van der Zwalmen P, Barlow P, Leroy F, Schoysman R: Application of partial zona dissection in human IVF. 7th World Congress on IVF and Assisted Procreation, 1991, Ab- str 108, p 145


Documents

Microsurgical fertilization procedures: The absence of stringent criteria for patient selection