Prevention of OHSS Shahar Kol IVF Unit Rambam

Prevention of OHSS Shahar Kol, IVF Unit Rambam Health Care Campus, and Macabbi Health Services, Haifa, Israel. February 2012

Content ● ● ● Scope of the problem Preventive strategies What really works Physiology of the agonist trigger Side benefits

Severe OHSS: is it still a problem? Maternal deaths and rates per 100, 000 ART procedures, including IVF: United Kingdom: 1997– 2005 Number Rate 95% CI Number of treatment cycles 1997– 1999 20 19. 17 12. 41– 29. 61 104, 320 2000– 2002 8 7. 32 3. 71– 14. 44 109, 308 2003– 2005 12 10. 08 5. 76– 17. 61 119, 080 Deaths Year * Source Human Fertilisation and Embryology Authority • “In 2003– 2005, 4 deaths (of the 12) were due to OHSS” • ~3 OHSS-related deaths per 100, 000 ART cycles

Three OHSS-related deaths (3: 100, 000), all had their embryos frozen Braat DDM, et al. Hum Reprod 2010; 25: 1782– 1786

Incidence and prediction of OHSS in women undergoing Gn. RH antagonist IVF cycles ● 2524 antagonist-based cycles (1801 patients) ● 53 patients (2%) were hospitalized because of OHSS – Conclusions: clinically significant OHSS is a limitation even in antagonist cycles “There is more than ever an urgent need for alternative final oocyte maturation – triggering medication” Papanikolaou EG, et al. Fertil Steril 2006; 85: 112– 120

Preventive strategies: coasting ● There was no evidence to suggest any benefit of withholding gonadotrophins (coasting) after ovulation in IVF for the prevention of OHSS D’angelo A, et al. Cochrane Database Syst Rev 2011; (6): CD 0028110

Preventive strategies: cryopreservation ● There is not enough evidence to show whether using frozen embryos …can reduce OHSS in women who are at high risk D’angelo A and Amso N. Cochrane Database Syst Rev 2007; (3): CD 002806

Preventive strategies: intravenous albumin ● Intravenous (iv) colloid fluids … at the time of oocyte retrieval may be beneficial for women with a high risk of developing OHSS ● Borderline evidence of benefit with the routine use of human albumin in the prevention of OHSS (1660 patients) ● Good evidence to support the use of hydroxyethyl starch in the prevention of OHSS (487 patients) Youssef MA, et al. Cochrane Database Syst Rev 2011; (2): CD 001302

IV albumin for the prevention of severe OHSS: a systemic review and meta-analysis ● 1199 patients ● IV albumin does not appear to reduce the occurrence of severe OHSS Venetis CA, et al. Fertil Steril 2011; 95: 188– 196, 196. e 1– 3

Preventive strategies: recombinant LH European Recombinant LH Study Group. J Clin Endocrinol Metab 2001; 86: 2607 – 2618

Treatment arm Parameters examined 5000 IU rh. LH (n=39) 15, 000 IU u-h. CG (n=34) 30, 000 IU 15, 000 + 10, 000 IU rh. LH (n=39) u-h. CG (n=41) rh. LH (n=26) u-h. CG (n=22) rh. LH (n=25) u-h. CG (n=24) 15. 17 ± 8. 34 15. 46 ± 6. 75 14. 23 ± 5. 61 14. 00 ± 4. 90 a a 0. 3007 11. 84 ± 7. 53 11. 78 ± 6. 75 12. 62 ± 6. 22 10. 82 ± 5. 70 a a 0. 1702 a a 0. 183 p (linearity) No. of follicles >10 mm 14. 03 ± 5. 32 16. 44 ± 6. 95 No. of oocytes retrieved 10. 23 ± 4. 70 11. 74 ± 6. 27 Oocytes in metaphase II 85. 5% 77. 8% No. of oocytes inseminated 9. 82 ± 4. 74 11. 26 ± 5. 73 11. 63 ± 7. 52 11. 57 ± 6. 57 12. 38 ± 6. 25 10. 55 ± 5. 74 a a 0. 1687 No. of embryos 5. 42 ± 3. 33 7. 00 ± 4. 68 6. 65 ± 5. 02 6. 36 ± 4. 68 7. 67 ± 4. 34 6. 33 ± 5. 19 a a 0. 0983 No. of embryos transferred 2. 39 ± 0. 60 2. 48 ± 0. 85 2. 58 ± 0. 6 2. 52 ± 0. 62 2. 78 ± 0. 8 2. 67 ± 0. 73 a a 0. 4310 Implantation rate 6. 0 ± 0. 16% 15. 0 ± 0. 31% 6. 0 ± 0. 19% 9. 0 ± 0. 24% 11. 0 ± 0. 26% 3. 0 ± 0. 09% 19. 0 ± 0. 33% 17. 0 ± 0. 33% 0. 1373 Pregnancy (total) 15. 4% (n=6) 26. 5% (n=9) 10. 3% (n=4) 24. 4% (n=10) 23. 1% (n=6) 13. 6% (n=3) 32. 0% (n=8) 37. 5% (n=9) 0. 2689 Clinical pregnancy 10. 3% (n=4) 23. 5% (n=8) 7. 7% (n=3) 14. 6% (n=6) 15. 4% (n=4) 13. 6% (n=3) 28. 0% (n=7) 25. 0% (n=6) 0. 1479 Live birth 5. 1% (n=2) 17. 6% (n=6) 7. 7% (n=3) 12. 2% (n=5) 15. 4% (n=4) 4. 5% (n=1) 20. 0% (n=5) 16. 7% (n=4) 0. 0606 Cryopreserved embryos 4. 42 ± 2. 65 6. 81 ± 3. 67 7. 93 ± 4. 18 4. 90 ± 3. 24 6. 27 ± 2. 96 4. 80 ± 3. 19 5. 75 ± 2. 49 9. 89 ± 3. 22 0. 2645 Cryopreserved embryos transferred 3. 42 ± 1. 83 5. 67 ± 2. 65 3. 50 ± 1. 84 3. 27 ± 1. 49 3. 00 ± 1. 41 2. 17 ± 0. 98 2. 50 ± 0. 71 4. 75 ± 2. 43 0. 9092 Pregnancy from cryopreserved embryos (total) 16. 7% (n=2/12) 0. 0% (n=0/9) 50. 0% (n=5/10) 27. 3% (n=3/11) 62. 5% (n=5/8) 33. 3% (n=2/6) 0. 0% (n=0/2) 0. 0% (n=0/8) b Clinical pregnancy from cryopreserved embryos 8. 3% (n=1/12) 0. 0% (n=0/9) 40. 0% (n=4/10) 27. 3% (n=3/11) 50. 0% (n=4/8) 16. 7% (n=1/6) 0. 0% (n=0/2) 0. 0% (n=0/8) b Live birth from cryopreserved embryos 8. 3% (n=1/12) 0. 0% (n=0/9) 30. 0% (n=3/10) 18. 2% (n=2/11) 12. 5% (n=1/8) 0. 0% (n=0/6) 0. 0% (n=0/2) 0. 0% (n=0/8) b a. The IVF data of days u-h. CG/rh. LH 0– 4 of patients from group 15, 000 90. 8% 88. 6% 57. 6% 84. 5% + 10, 000 IU were pooled with those from group 15, 000 IU b. Because the numbers were small, no statistical comparison was performed on these data ● 15, 000 + 10, 000 IU gave 20% live birth rate but with a 12% OHSS rate European Recombinant LH Study Group. J Clin Endocrinol Metab 2001; 86: 2607 – 2618

Preventive strategies: lowering h. CG dose ● Reducing the dose of h. CG does not eliminate the risk of OHSS in a high-risk group Schmidt DW, et al. Fertil Steril 2004; 82(4): 841– 846

Preventive strategies: dopamine agonists OHSS incidence OHSS severity Youssef MA, et al. Human Reprod Update 2010; 16: 459– 466

What really works: ● Gn. RH agonist versus h. CG for oocyte triggering in Gn. RH antagonist ART cycles Youssef MA, et al. Human Reprod Update 2010; 16: 459– 466

Ovulation trigger RCT, high risk Oocyte source Own Engamnn, et al 2008 RCT, high risk Own Gn. RHa h. CG Acevedo, et al 2006 RCT Donors Gn. RHa h. CG Bodri, et al 2009 Retrospective Donors Gn. RHa h. CG Griesinger, et al 2010 Observational, High risk RCT Own Gn. RHa Own Retrospective, case- Own controlled, high risk Reference Trial type Babayof, et al 2006 • 16 publications • Agonist: 2005 patients, not a single Humaidan, et al 2009 case of OHSS! Engmann, et al 2006 • h. CG: 92 cases in 1810 patients, 5. 1% Manzanares, et al 2009 Retrospective case- Own control, high risk n OHSS % (n) 15 13 33 32 30 30 1046 1031 0 (0/13) 31(4/13) 0 (0/33) 31 (10/32) 0 (0/30) 17 (5/30) 0 (0/1046) 1. 3 (13/1031) 40 0 (0/40) Gn. RHa h. CG 152 150 0 (0/152) 2 (3/150) Gn. RHa h. CG 23 23 0 (0/23) 4 (1/23) 42 0 (0/42) 254 175 82 69 0 (0/254) 6 (10/175) 0 (0/82) 7 (5/69) 32 42 0 (0/32) 1 (1/42) 44 44 0 (0/44) 7 (3/44) 12 8 (1/12) 106 50 50 4 45 0 (0/106) 8 (9/106) 0 (0/50) 16(8/50) 0 (0/45) 15 (33) Gn. RHa h. CG - cancelled Hernandez, et al 2009 Retrospective Donors Gn. RHa h. CG Orvieto, et al 2006 Retrospective, high risk: agonist arm only RCT Own Gn. RHa h. CG Donors Gn. RHa h. CG Shapiro, et al 2007 Sismanoglu, et al 2009 Gn. RH, luteal rescue with h. CG 1500 IU Gn. RHa h. CG Humaidan, et al 2009 Observational, high Own risk Galindo, et al 2009 RCT Donors Melo, et al 2009 RCT Donors Gn. RHa h. CG Shahrokh, et al 2010 RCT, high risk Own Gn. RHa h. CG

Failures? OHSS prevention by Gn. RH agonist triggering of final oocyte maturation in a Gn. RH antagonist protocol in combination with freeze-all strategy: a prospective multicenter study ● Conclusions: “…a single case of a severe early onset OHSS occurred” – E 2 trigger day=47, 877 pmol/L – 13 oocytes – “drastic decrease of hemoglobin levels to 4. 9 mmol/L” (8 grams/d. L) patient received blood transfusion 2 days post OPU – Hematocrit: 41 trigger day, 37 OPU day, ‘, <35’ post blood transfusion – 3– 4 days post trigger 3. 9 litres of “blood-stained ascites which was indicative of a subacute intraperitoneal hemorrhage” Griesinger G, et al. Fertil Steril 2011; 95: 2029– 2033

The physiology of agonist trigger LH surge 1 1. 2. FSH surge 2 Humaidan P, et al. Reprod Biomed Online 2011; (Epub ahead of print); Gonen Y, et al. J Clin Endocrinol Metab 1990; 71: 918– 922

What happens after agonist trigger? Complete luteolysis! Luteal phase Natural cycle Day 7– 9 = 75 pg/m. L vs 18 Natural cycle Day 7– 9 = 750 pg/m. L vs 84 Nevo O, et al. Fertil Steril 2003; 79: 1123– 1128

How to secure good clinical outcome post agonist trigger? ● High risk fresh transfer: intensive E 2+P luteal support ● High risk: ‘freeze-all’ ● Low risk: luteal rescue based on LH activity

Luteal phase: intensive E+P OHSS high-risk patients Study group Control group Odds ratio (95%CI) p value Total, n (%) 0/33 (0) 10/32 (31. 3) 0 (0– 0. 26)a <0. 01 Moderate/severe, n (%) 0/33 (0) 5/32 (15. 6) 0 (0– 0. 74)a 0. 02 Total, n (%) 0/30 (0) 10/2 (34. 5) 0 (0– 0. 26)a <0. 01 Moderate/severe, n (%) 0/30 (0) 5/29 (17. 2) 0 (0– 0. 73)a 0. 02 22/61 (36) 20/64 (31) 1. 18 (0. 52– 2. 65) 0. 69 Positive pregnancy, n (%) 19/30 (63. 3) 18/29 (62. 1) 1. 06 (0. 37– 3. 0) 0. 92 Clinical pregnancy rate, n (%) 17/30 (56. 7) 15/29 (51. 7) 1. 22 (0. 4– 3. 4) 0. 45 Ongoing pregnancy rate, n (%) 16/30 (53. 3) 14/29 (48. 3) 1. 22 (0. 4– 3. 4) 0. 45 Primary end points OHSS (ITT) OHSS (PP) Secondary end point (PP) Implantation rate, n (%) Other end points (PP) a. The estimates of these odds ratios are zero, because no patient developed OHSS in the study group; ITT=intention to treat; PP=per protocol Engmann L, et al. Fertil Steril 2008; 89: 84– 91

Modified luteal support post agonist trigger 1500 IU h. CG administered at oocyte retrieval rescues the luteal phase when Gn. RH agonist is used for ovulation induction: a prospective, randomized, controlled study ● 305 patients ● No significant differences were seen regarding: – – – Positive h. CG/ET rate (48 and 48%) Ongoing pregnancy rate (26 and 33%) Delivery rate (24 and 31%) Rate of early pregnancy loss (21 and 17%) Between the Gn. RHa and 10, 000 intrauterine h. CG groups, respectively Humaidan P, et al. Fertil Steril 2010; 93: 847– 854

Tailored luteal phase support Patients with ≤ 14 follicles ≥ 12 mm on day of trigger Gn. RHa + 1500 IU h. CG x 2, versus 5000 IU h. CG, both groups E 2+P luteal support. Gn. RHa/h. CG Patients, n h. CG 125 141 110/125 (88) 116/141 (82) 1. 3 IR 49/158 (36) 43/145 (30) Pos h. CG per ET, n (%) 47/110 (43) 41/116 (35) Clinical pregnancy per patient, n (%) 43/125 (34) 40/141 (28) Ongoing pregnancy per patient, n (%) 37/125 (30) 36/141 (26) Rate of transfer, n (%) Embryos transferred, mean Humaidan P, et al. personal communication

Side benefits ● Agonist trigger: more MII oocytes compared with h. CG trigger 1 -4 ● Potential benefit of FSH surge: 5 -9 – Promotes LH receptor formation in luteinizing granulosa cells – Promotes nuclear maturation (i. e. resumption of meiosis) – Promotes cumulus expansion 1. 2. 3. 4. 5. 6. 7. 8. 9. Humaidan P, et al. Reprod Biomed Online 2005; 11: 679– 684 Humaidan P, et al. Human Reprod 2009; 24: 2389– 2394 Imoedemhe DA, et al. Fertil Steril 1991; 55: 328– 332 Oktay K, et al. Reprod Biomed Online 2010; 20: 783– 788 Eppig JJ. Nature 1979; 281: 483– 484 Strickland Beers. J Biol Chem 1976; 251: 5694– 5702 Yding Andersen C. Reprod Biomed Online 2002; 5: 232– 239 Yding Andersen C, et al. Mol Hum Reprod 1999; 5: 726– 731 Zelinski-Wooten MB, et al. Human Reprod 1995; 10: 1658– 1666

The advantage for the ‘normal responder’ Antagonist FSH/h. MG Agonist trigger OPU 36 hours ET 4 days 1500 IU h. CG Kol S, et al. Human Reprod 2011; 26: 2874– 2877 1500 IU h. CG

Stimulation characteristics and embryology data Stimulation (days) 9. 3 ± 2. 0 Gn. RH antagonist (days) 3. 8 ± 0. 9 FSH (units) 2443 ± 925 E 2 day of trigger (pmol/L) 3764 ± 1227 P day of trigger (nmol/L) 2. 4 ± 1. 65 LH day of trigger (IU/L) 1. 9 ± 1. 3 Oocytes retrieved 6. 7 ± 2. 5 Embryos obtained 3. 6 ± 1. 7 Embryos transferred 2. 9 ± 0. 9 Embryos frozen 0. 8 ± 1. 5 Beta h. CG (IU/L) 152 ± 86 E 2 (day of pregnancy test, pmol/L) 6607 ± 3789 P (day of pregnancy test, nmol/L) 182 ± 50 Values are mean ± SD Reproductive outcomes Positive h. CG/cycle, n (%) 11/15 (73) Clinical ongoing pregnancy, n (%) 7/15 (47) Early pregnancy loss, n (%) 4/11 (36) Kol S, et al. Human Reprod 2011; 26: 2874– 2877

“The concept of an OHSS-Free Clinic has become a reality. This approach should include pituitary down-regulation using a Gn. RH antagonist, ovulation triggering with a Gn. RH agonist and vitrification of oocytes or embryos” “…luteal phase supplementation with low-dose h. CG has to be fine tuned. ” Devroey P, et al. Human Reprod 2011; 26: 2593– 2597

Crystal ball: where are we heading? In Out Antagonist-based protocols ‘Long agonist’ protocols Agonist trigger h. CG trigger LH activity-based luteal support Progesterone-based luteal support Total OHSS elimination 1– 2% severe OHSS Total OHSS elimination OHSS-related death rate: 3: 100, 000 Patient-friendly luteal phase Painful P injections or leaky, messy vaginal P Thank you