- Email: [email protected]
Microsurgical ovarian transplantation in the primate
FERTILITY AND STERILITY Copyright c 1981 The American Fertility Society
Vol. 36, No.4, October 1981
Printed in U.SA.
MICROSURGICAL OVARIAN TRANSPLANTATION IN THE PRIMATE
JAMES R. SCO'IT, M.D.* WILLIAM R. KEYE, M.D. A. MARSH POULSON, M.D. W. ANN REYNOLDS, PH.D. Department of Obstetrics. and Gynecology, University of Utah Medical Center, Salt Lake City, Utah 84131
We have developed a technique for orthotopic transplantation of ovaries by microsurgical reanastomosis of the ovarian blood vessels in rhesus monkeys. Four monkeys receiving autografts resumed cyclic menses and had postoperative circulating luteinizing hormone (LH) and progesterone concentrations consistent with developing follicles and corpus luteum function. Postoperative luteal phase ovarian biopsies were indicative of ovulation in three of the four animals. However, in a fifth recipient an ovarian allograft from an unrelated donor was rejected despite the use of a standard immunosuppressive regimen. This study suggests that ovarian transplantation by direct vascular anastomosis is a technically feasible surgical procedure in the human, where the vascular anatomy is similar and the caliber of the ovarian vessels is greater than in the rhesus monkey. Fertil Steril 36:512, 1981
Ovarian transplantation between monozygotic twins already has therapeutic possibilities, and the procedure may eventually be clinically useful in immunogenetically dissimilar individuals. However, unless direct vascular anastomosis can be accomplished, the ovary that is unable to recreate its specialized cells would inevitably suffer from the consequences of ischemia before normal circulation was restored. We describe here a technique for orthotopic transplantation of ovaries by microsurgical reanastomosis of the ovarian artery and vein in previously oophorectomized rhesus monkeys. MATERIALS AND METHODS
Animals. Five mature rhesus monkeys with a history of normal menstrual cycles were housed under conditions of controlled temperature and Received March 20, 1981; revised and accepted June 26, 1981. *Reprint requests: James R. Scott, M.D., Professor and Chairman, Department of Obstetrics and Gynecology, University of Utah Medical Center, Salt Lake City, Utah 84132.
humidity, with exposure to 14 hours of fluorescent light per day. Each monkey underwent laparotomy through a lower abdominal midline incision under halothane anesthesia. A left oophorectomy was performed in all animals by excision of the ovary after double ligation of the infundibulopelvic ligament with 4-0 chromic catgut suture. In the four monkeys that received an autograft, the right ovarian artery was dissected free to its origin at the aorta, where we removed it by taking a small eliptical patch of the aortic wall. The right ovarian vein was dissected free to the vena cava in a similar manner and removed with a small patch of vena caval wall. After flushing the vessels with a dilute heparin solution, we performed reanastomosis under the operating microscope, using 8-0 nylon running sutures. In two animals, the right ovarian artery was resutured to the aorta at its site of origin and the right ovarian vein was replaced where it had been removed from the vena cava (Fig. 1 A). In two animals, the right ovarian artery was implanted into the right common iliac artery and the right ovarian vein was implanted into the right common iliac vein (Fig. 1 B). Mean ischemia time before
512
Vol. 36, No.4
513
OVARIAN TRANSPLANTATION IN THE PRIMATE
been described in detail. 1 The standard employed in the LH radioimmunoassay was a partially purified preparation of monkey pituitary gonadotropin (LER-M-907-D) that had an LH potency of 0.025 NIH-LH-Sl units/mg (ovarian ascorbic acid depletion assay). RESULTS
FIG. 1. A, Ovarian artery reimplanted into aorta and ovarian vein reimplanted into vena cava. B, Ovarian artery implanted into right common iliac artery and ovarian vein implanted into right common iliac vein.
complete reanastomosis of the ovarian blood supply was approximately 1 hour. The fifth monkey received an ovarian allograft from an unrelated donor following a bilateral oophorectomy. The ovarian vessels were implanted into the right common iliac vessels. Oral azathioprine, 4 mg/kg/day, and intramuscular methylprednisolone, 0.5 mg/kg/day, were begun the day prior to surgery and continued postoperatively. In all animals peritoneal defects were closed with running 6-0 chromic catgut sutures. In each case the ovary was placed in its natural position in the right side of the pelvis, anchored to the fimbriated end of the right fallopian tube with one 6-0 chromic suture. The abdomen was closed in a conventional manner, and each monkey received 400,000 units of penicillin and 0.5 gm of streptomycin on the day of surgery. Menstrual records were kept, and blood samples were obtained postoperatively from conscious animals by puncture of the femoral artery or vein between 0800 and 1000 hours every one to four days for at least one menstrual cycle in each monkey. After normal menses had resumed, each ovarian autograft recipient was mated with a mature rhesus male from day 10 to day 14 of the menstrual cycle. Second-look laparotomies and ovarian biopsies were also carried out at varying intervals after the original surgery. Assay Procedures. The radioimmunoassay procedures utilized for the determination of monkey luteinizing hormone (LH) and progesterone have
All animals tolerated the surgical procedure well. Unfortunately, one monkey (3418) died 7 months postoperatively of acute gastric dilatation of undetermined cause. Each ovarian autograft recipient had scant vaginal bleeding from 1 to 2 days following surgery, the onset of a menstrual period 4 to 8 weeks postoperatively, and resumption of relatively normal cyclic menses thereafter. The mean duration of menstrual cycles for all animals with ovarian autograftspostoperatively was 30.9 days (Table 1), which was a pattern similar to their preoperative patterns and cycle length in normal rhesus monkeys. 2 The ovarian allograft recipient remained amenorrheic postoperatively. Circulating LH and luteal phase progesterone concentrations, appearance of the ovaries at second-look laparotomies, and histologic appearance of ovarian biopsies (Table 2 and Fig. 2) suggested normal endocrine function and ovulatory cycles in three of the four animals with ovarian autografts. In contrast, the postoperative serum LH concentrations from the ovarian allograft recipient were in the postmenopausal range, serum progesterone levels were undetectable, and there was no evidence of viable ovarian tissue at laparotomy 2 months after the original surgery. Mild to moderate tuboovarian adhesions, which had formed in each instance, were lysed during the second-look operations. Although each of the monkeys with ovarian autografts was placed with a male for at least 6 menstrual cycles, no pregnancies were achieved. DISCUSSION
The ovary presents a challenge in transplantation because of its particular vascular supply, inTABLE 1. Rhesus Monkey Menstrual Cycle Length in Days Before and After Ovarian Autotransplantation Duration of menstrual cycles
Reported normal 2 Preoperative Postoperative
Mean ± SEM
Range
27.5 ± 0.7 (N = 60) 22-41 (N = 60) 28.1 ± 0.8 (N = 7) 26-31 (N = 7) 30.9 ± 1.9 (N = 23) 17-60 (N = 23)
SCOTI' ET AL.
514
October 1981
TABLE 2. Postoperative Laboratory Data on Rhesus Monkeys Undergoing Ovarian Transplantation LH
Animal
Procedure
(mean ~ SEM)
fJlflml
3640 3969 3970 3418 3399 3113
Unoperated control 2.19 Autotransplant 0.74 Autotransplant 1.50 Autotransplant 3.94 Autotransplant 1.28 Allograft; 23.81
Progesterone (mean Follicular
SEM)
Ovarian histology
Luteal
fJlf1ml
0.34 (n = 20) 0.25 ± 0.11 (n = 0.16 (n = 21) 0.20 ± 0.08 (n = 0.51 (n = 17) 0.30 ± 0.13 (n = 0.52 (n = 7) 0.36 ± 0.07 (n = 0.29 (n = 13) 0.18 ± 0.04 (n = ± 2.43 (n = 7) ND b (n = 3)
± ± ± ± ±
~
Graafian follicles
Corpus luteum
+ + + +
+
fJlflml
8) 4) 3) 4) 4)
5.64 ± 0.72 (n 5.90 ± 2.02 (n
= 9) = 3)
0.48 ± 0.18 (n = 3) 1.08 ± 0.62 (n = 9) ND (n = 4)
+
+
"Insufficient serum to perform determinations. bND = not detectable.
nervation, and lymphatic drainage. Experimental studies are necessary for one to establish a reliable and safe surgical procedure and determine whether normal endocrine function can be maintained. Although ovarian transplantation by microvascular reanastomosis techniques has been successfully accomplished in lower animals,3-5 it is difficult to relate the results in those species to the clinical situation in man. Therefore, the rhesus monkey was selected for this study because the reproductive tract closely resembles, anatomically and physiologically, that of the human female. Our results indicate that the surgical procedure is technically possible, and there is a reasonable chance of maintaining ovarian endocrine function. Evidence for ovulation in at least one postoperative menstrual cycle was obtained by luteal phase progesterone concentrations in one animal and by histologic demonstration of corpora lutea in the autotransplanted ovaries of three animals. Unfortunately, this procedure may not be as successful in restoring fertility, for the results of this study suggest that postoperative tuboovarian ad-
FIG. 2. Biopsy specimen of autotransplanted ovary from monkey 3399 six months postoperatively illustrating corpus luteum (CL) adjacent to ovarian cortex containing follicles (F).
hesions may represent a significant mechanical deterrent to the achievement of pregnancy. Perhaps one of the anti adhesion regimens commonly employed with tubal surgery could successfully inhibit adhesion formation. 6 Until the problems of allograft rejection are overcome, transplantation of nonvital organs will not become a reality. In general, the pattern of rejection of the ovary is similar to that of other allografted organs. 7 However, ovarian transplantation between immunologically compatible individuals such as monozygotic twins is a potentially useful procedure where one sister has been surgically castrated or has primary ovarian failure, or perhaps when one twin is chromosomally normal and the other has gonadal dysgenesis. 8 In fact, testicular transplantation from an identical twin to his anorchic brother by microvascular anastomosis of the spermatic vessels has already been successfully utilized to restore normal male fertility.9 In addition, in certain infertile women with one absent ovary, transplantation of the remaining ovary to the opposite side of the pelvis might be considered with an absent or blocked ipsilateral fallopian tube and a normal contralateral fallopian tube. Since the ovarian artery and vein are larger in diameter than they are in the rhesus monkey, ovarian transplantation by direct vascular anastomosis should be technically easier to perform in the human. Although this operation has to be considered experimental at present and requires further evaluation and refinement, ovarian transplantation may well become an accepted surgical procedure for selected patients in the future.
Acknowledgments. The authors are indebted to Dr. Gordon Niswender for providing the antiserum used for LH determinations, to Steve Kelley for his expert surgical assistance, to Jan Corey for collecting blood samples, and to Joanne West for performing the radioimmunoassays.
Vol. 36, No.4
OVARIAN TRANSPLANTATION IN THE PRIMATE
REFERENCES 1. Hodgen GD, WilksJW, VaitukaitisJL, Chen HC, Papkoff
H, Ross GT: A new radioimmunoassay for follicle stimulating hormone in macaques: ovulatory menstrual cycles. Endocrinology 99:137, 1976 2. Wilks JW, Hodgen GD, Ross GT: Luteal phase defects in the rhesus monkey: the significance of serum FSH:LH ratios. J Clin Endocrinol Metab 43:1261,1976 3. Schmidt FL, Miller RL, Peterson T, Soret MG: A microsurgical technique for orthotopic ovarian transplantation in the Chinese hamster. Surgery 80:595,1976 4. Betteridge KG: Homotransplantation of ovaries with vascular anastomoses in rabbits: response of transplants to human chorionic gonadotropin. J EndocrinoI47:451, 1970
515
5. Baird DT, Land RB, Scaramuzzi RJ, Wheeler AG: Functional assessment of the autotransplanted uterus and ovary in the ewe. Proc Roy Soc Lond [B] 192:463, 1976 6. Pfeffer WH: Adjuvants in tubal surgery. Fertil Steril 33:245, 1980 7. Mattingly RF, Clark DO, Lutsky II, Huang WY, Staff A, Maddison FE: Ovarian function in uteroovarian homotransplantation. Am J Obstet Gynecol 108:773, 1970 8. Ross GT, Tjio JH, Lipsett MB: Cytogenetic studies of presumptively monozygotic twin girls discordant for gonadal dysgenesis. J Clin Endocrinol Metab 29:440, 1969 9. Silber SS, Rodriguez-Rigau LJ: Pregnancy after testicular transplant: importance of treating the couple. Fertil Steril 33:454, 1980
Vol. 36, No.4, October 1981
Printed in U.SA.
MICROSURGICAL OVARIAN TRANSPLANTATION IN THE PRIMATE
JAMES R. SCO'IT, M.D.* WILLIAM R. KEYE, M.D. A. MARSH POULSON, M.D. W. ANN REYNOLDS, PH.D. Department of Obstetrics. and Gynecology, University of Utah Medical Center, Salt Lake City, Utah 84131
We have developed a technique for orthotopic transplantation of ovaries by microsurgical reanastomosis of the ovarian blood vessels in rhesus monkeys. Four monkeys receiving autografts resumed cyclic menses and had postoperative circulating luteinizing hormone (LH) and progesterone concentrations consistent with developing follicles and corpus luteum function. Postoperative luteal phase ovarian biopsies were indicative of ovulation in three of the four animals. However, in a fifth recipient an ovarian allograft from an unrelated donor was rejected despite the use of a standard immunosuppressive regimen. This study suggests that ovarian transplantation by direct vascular anastomosis is a technically feasible surgical procedure in the human, where the vascular anatomy is similar and the caliber of the ovarian vessels is greater than in the rhesus monkey. Fertil Steril 36:512, 1981
Ovarian transplantation between monozygotic twins already has therapeutic possibilities, and the procedure may eventually be clinically useful in immunogenetically dissimilar individuals. However, unless direct vascular anastomosis can be accomplished, the ovary that is unable to recreate its specialized cells would inevitably suffer from the consequences of ischemia before normal circulation was restored. We describe here a technique for orthotopic transplantation of ovaries by microsurgical reanastomosis of the ovarian artery and vein in previously oophorectomized rhesus monkeys. MATERIALS AND METHODS
Animals. Five mature rhesus monkeys with a history of normal menstrual cycles were housed under conditions of controlled temperature and Received March 20, 1981; revised and accepted June 26, 1981. *Reprint requests: James R. Scott, M.D., Professor and Chairman, Department of Obstetrics and Gynecology, University of Utah Medical Center, Salt Lake City, Utah 84132.
humidity, with exposure to 14 hours of fluorescent light per day. Each monkey underwent laparotomy through a lower abdominal midline incision under halothane anesthesia. A left oophorectomy was performed in all animals by excision of the ovary after double ligation of the infundibulopelvic ligament with 4-0 chromic catgut suture. In the four monkeys that received an autograft, the right ovarian artery was dissected free to its origin at the aorta, where we removed it by taking a small eliptical patch of the aortic wall. The right ovarian vein was dissected free to the vena cava in a similar manner and removed with a small patch of vena caval wall. After flushing the vessels with a dilute heparin solution, we performed reanastomosis under the operating microscope, using 8-0 nylon running sutures. In two animals, the right ovarian artery was resutured to the aorta at its site of origin and the right ovarian vein was replaced where it had been removed from the vena cava (Fig. 1 A). In two animals, the right ovarian artery was implanted into the right common iliac artery and the right ovarian vein was implanted into the right common iliac vein (Fig. 1 B). Mean ischemia time before
512
Vol. 36, No.4
513
OVARIAN TRANSPLANTATION IN THE PRIMATE
been described in detail. 1 The standard employed in the LH radioimmunoassay was a partially purified preparation of monkey pituitary gonadotropin (LER-M-907-D) that had an LH potency of 0.025 NIH-LH-Sl units/mg (ovarian ascorbic acid depletion assay). RESULTS
FIG. 1. A, Ovarian artery reimplanted into aorta and ovarian vein reimplanted into vena cava. B, Ovarian artery implanted into right common iliac artery and ovarian vein implanted into right common iliac vein.
complete reanastomosis of the ovarian blood supply was approximately 1 hour. The fifth monkey received an ovarian allograft from an unrelated donor following a bilateral oophorectomy. The ovarian vessels were implanted into the right common iliac vessels. Oral azathioprine, 4 mg/kg/day, and intramuscular methylprednisolone, 0.5 mg/kg/day, were begun the day prior to surgery and continued postoperatively. In all animals peritoneal defects were closed with running 6-0 chromic catgut sutures. In each case the ovary was placed in its natural position in the right side of the pelvis, anchored to the fimbriated end of the right fallopian tube with one 6-0 chromic suture. The abdomen was closed in a conventional manner, and each monkey received 400,000 units of penicillin and 0.5 gm of streptomycin on the day of surgery. Menstrual records were kept, and blood samples were obtained postoperatively from conscious animals by puncture of the femoral artery or vein between 0800 and 1000 hours every one to four days for at least one menstrual cycle in each monkey. After normal menses had resumed, each ovarian autograft recipient was mated with a mature rhesus male from day 10 to day 14 of the menstrual cycle. Second-look laparotomies and ovarian biopsies were also carried out at varying intervals after the original surgery. Assay Procedures. The radioimmunoassay procedures utilized for the determination of monkey luteinizing hormone (LH) and progesterone have
All animals tolerated the surgical procedure well. Unfortunately, one monkey (3418) died 7 months postoperatively of acute gastric dilatation of undetermined cause. Each ovarian autograft recipient had scant vaginal bleeding from 1 to 2 days following surgery, the onset of a menstrual period 4 to 8 weeks postoperatively, and resumption of relatively normal cyclic menses thereafter. The mean duration of menstrual cycles for all animals with ovarian autograftspostoperatively was 30.9 days (Table 1), which was a pattern similar to their preoperative patterns and cycle length in normal rhesus monkeys. 2 The ovarian allograft recipient remained amenorrheic postoperatively. Circulating LH and luteal phase progesterone concentrations, appearance of the ovaries at second-look laparotomies, and histologic appearance of ovarian biopsies (Table 2 and Fig. 2) suggested normal endocrine function and ovulatory cycles in three of the four animals with ovarian autografts. In contrast, the postoperative serum LH concentrations from the ovarian allograft recipient were in the postmenopausal range, serum progesterone levels were undetectable, and there was no evidence of viable ovarian tissue at laparotomy 2 months after the original surgery. Mild to moderate tuboovarian adhesions, which had formed in each instance, were lysed during the second-look operations. Although each of the monkeys with ovarian autografts was placed with a male for at least 6 menstrual cycles, no pregnancies were achieved. DISCUSSION
The ovary presents a challenge in transplantation because of its particular vascular supply, inTABLE 1. Rhesus Monkey Menstrual Cycle Length in Days Before and After Ovarian Autotransplantation Duration of menstrual cycles
Reported normal 2 Preoperative Postoperative
Mean ± SEM
Range
27.5 ± 0.7 (N = 60) 22-41 (N = 60) 28.1 ± 0.8 (N = 7) 26-31 (N = 7) 30.9 ± 1.9 (N = 23) 17-60 (N = 23)
SCOTI' ET AL.
514
October 1981
TABLE 2. Postoperative Laboratory Data on Rhesus Monkeys Undergoing Ovarian Transplantation LH
Animal
Procedure
(mean ~ SEM)
fJlflml
3640 3969 3970 3418 3399 3113
Unoperated control 2.19 Autotransplant 0.74 Autotransplant 1.50 Autotransplant 3.94 Autotransplant 1.28 Allograft; 23.81
Progesterone (mean Follicular
SEM)
Ovarian histology
Luteal
fJlf1ml
0.34 (n = 20) 0.25 ± 0.11 (n = 0.16 (n = 21) 0.20 ± 0.08 (n = 0.51 (n = 17) 0.30 ± 0.13 (n = 0.52 (n = 7) 0.36 ± 0.07 (n = 0.29 (n = 13) 0.18 ± 0.04 (n = ± 2.43 (n = 7) ND b (n = 3)
± ± ± ± ±
~
Graafian follicles
Corpus luteum
+ + + +
+
fJlflml
8) 4) 3) 4) 4)
5.64 ± 0.72 (n 5.90 ± 2.02 (n
= 9) = 3)
0.48 ± 0.18 (n = 3) 1.08 ± 0.62 (n = 9) ND (n = 4)
+
+
"Insufficient serum to perform determinations. bND = not detectable.
nervation, and lymphatic drainage. Experimental studies are necessary for one to establish a reliable and safe surgical procedure and determine whether normal endocrine function can be maintained. Although ovarian transplantation by microvascular reanastomosis techniques has been successfully accomplished in lower animals,3-5 it is difficult to relate the results in those species to the clinical situation in man. Therefore, the rhesus monkey was selected for this study because the reproductive tract closely resembles, anatomically and physiologically, that of the human female. Our results indicate that the surgical procedure is technically possible, and there is a reasonable chance of maintaining ovarian endocrine function. Evidence for ovulation in at least one postoperative menstrual cycle was obtained by luteal phase progesterone concentrations in one animal and by histologic demonstration of corpora lutea in the autotransplanted ovaries of three animals. Unfortunately, this procedure may not be as successful in restoring fertility, for the results of this study suggest that postoperative tuboovarian ad-
FIG. 2. Biopsy specimen of autotransplanted ovary from monkey 3399 six months postoperatively illustrating corpus luteum (CL) adjacent to ovarian cortex containing follicles (F).
hesions may represent a significant mechanical deterrent to the achievement of pregnancy. Perhaps one of the anti adhesion regimens commonly employed with tubal surgery could successfully inhibit adhesion formation. 6 Until the problems of allograft rejection are overcome, transplantation of nonvital organs will not become a reality. In general, the pattern of rejection of the ovary is similar to that of other allografted organs. 7 However, ovarian transplantation between immunologically compatible individuals such as monozygotic twins is a potentially useful procedure where one sister has been surgically castrated or has primary ovarian failure, or perhaps when one twin is chromosomally normal and the other has gonadal dysgenesis. 8 In fact, testicular transplantation from an identical twin to his anorchic brother by microvascular anastomosis of the spermatic vessels has already been successfully utilized to restore normal male fertility.9 In addition, in certain infertile women with one absent ovary, transplantation of the remaining ovary to the opposite side of the pelvis might be considered with an absent or blocked ipsilateral fallopian tube and a normal contralateral fallopian tube. Since the ovarian artery and vein are larger in diameter than they are in the rhesus monkey, ovarian transplantation by direct vascular anastomosis should be technically easier to perform in the human. Although this operation has to be considered experimental at present and requires further evaluation and refinement, ovarian transplantation may well become an accepted surgical procedure for selected patients in the future.
Acknowledgments. The authors are indebted to Dr. Gordon Niswender for providing the antiserum used for LH determinations, to Steve Kelley for his expert surgical assistance, to Jan Corey for collecting blood samples, and to Joanne West for performing the radioimmunoassays.
Vol. 36, No.4
OVARIAN TRANSPLANTATION IN THE PRIMATE
REFERENCES 1. Hodgen GD, WilksJW, VaitukaitisJL, Chen HC, Papkoff
H, Ross GT: A new radioimmunoassay for follicle stimulating hormone in macaques: ovulatory menstrual cycles. Endocrinology 99:137, 1976 2. Wilks JW, Hodgen GD, Ross GT: Luteal phase defects in the rhesus monkey: the significance of serum FSH:LH ratios. J Clin Endocrinol Metab 43:1261,1976 3. Schmidt FL, Miller RL, Peterson T, Soret MG: A microsurgical technique for orthotopic ovarian transplantation in the Chinese hamster. Surgery 80:595,1976 4. Betteridge KG: Homotransplantation of ovaries with vascular anastomoses in rabbits: response of transplants to human chorionic gonadotropin. J EndocrinoI47:451, 1970
515
5. Baird DT, Land RB, Scaramuzzi RJ, Wheeler AG: Functional assessment of the autotransplanted uterus and ovary in the ewe. Proc Roy Soc Lond [B] 192:463, 1976 6. Pfeffer WH: Adjuvants in tubal surgery. Fertil Steril 33:245, 1980 7. Mattingly RF, Clark DO, Lutsky II, Huang WY, Staff A, Maddison FE: Ovarian function in uteroovarian homotransplantation. Am J Obstet Gynecol 108:773, 1970 8. Ross GT, Tjio JH, Lipsett MB: Cytogenetic studies of presumptively monozygotic twin girls discordant for gonadal dysgenesis. J Clin Endocrinol Metab 29:440, 1969 9. Silber SS, Rodriguez-Rigau LJ: Pregnancy after testicular transplant: importance of treating the couple. Fertil Steril 33:454, 1980