Aspirin

No aspirin

p

Number patients

72 (23%)

244 (77%)

Age

34 ± 4

34 ± 4

0.7

Previously pregnant

7 (10%)

39 (16%)

0.06

Diagnosis

• Unexplained

4 (6%)

12 (5%)

• Male factor

23 (32%)

51 (21%)

• Endometriosis

9 (13%)

51 (21%)

• Tubal factor

13 (18%)

54 (22%)

• Ovulatory dysfunction

12 (17%)

24 (10%)

• Other

6 (8%)

17 (7%)

• Multiple diagnoses

4 (6%)

54 (22%)

Prior IVF

32%

22%

0.12

Basal FSH (mIU/mL)

7 ± 2

7 ± 8

0.14

Stimulation method

• Antagonist

0

2 (1%)

• Long luteal leuprolide

62 (86%)

195 (80%)

• Flare

8 (11%)

34 (14%)

Low-dose aspirin did not improve any IVF outcomes analyzed in this study, even though more embryos were transferred to women who used aspirin (p = 0.03) (Table 2). In fact, the pregnancy rate in aspirin users was 48%, slightly lower compared to non-users, 57% (p = 0.18). Clinical pregnancy rates were 45% and 54% for users and non-users, respectively. Live birth rates tended to be lower with aspirin, 29% and 41%, respectively (p = 0.07). Implantation rates were significantly lower in patients who received aspirin, 21% and 30%, respectively (p = 0.01). Maximal endometrial thickness was not improved with aspirin compared to non-aspirin controls (p = 0.26). The percentage of ICSI cycles was similar in each group, as was the number of eggs fertilized.

Low-dose aspirin did not benefit IVF patients in our program. Aspirin therapy did not enhance endometrial thickness, augment the ovarian response, or improve pregnancy rates. The demographics were similar between the two groups of patients, with similarities in diagnosis, stimulation protocol, as well as number of ICSI cycles.

Our results conflict with several studies that have shown that aspirin is beneficial for infertility therapy. Rubenstein et al found that aspirin 100 mg starting in the luteal phase of the preceding cycle improved blood flow velocity, ovarian responsiveness, implantation and pregnancy rates in a randomized, controlled trial of 149 patients undergoing IVF compared to 149 placebo-treated controls 25. Weckstein et al also found enhanced uterine blood flow and significantly higher implantation and clinical pregnancy rates with low-dose aspirin in women who had a thin endometrium undergoing embryo transfer from oocyte donation in a randomized controlled study 3.

Interestingly, endometrial thickness was not improved with aspirin. In an prospective, randomized insemination study of women with a thin endometrium undergoing insemination, aspirin improved the percentage of trilaminar endometrium and pregnancy rates from 9 to 18%, but not endometrial thickness or ultrasound flow patterns 6.

Waldenstrom et al randomized 1380 unselected IVF cycles on alternate days to receive aspirin 75 mg or no aspirin starting on the day of embryo transfer and continuing until 18 days after retrieval 7. In this study, the live birth rate was 27% with aspirin and 23% in the control population, with an odds ratio 1.2 (95% CI 1.0–1.6). A non-controlled study found that IVF outcome was significantly improved when aspirin, heparin, and intravenous immunoglobulin therapy was administered to women with repeat IVF failures and antiphospholipid antibodies, but not to women with negative antiphospholipid antibodies 8. Other studies have also found a beneficial effect with aspirin/heparin, and aspirin plus prednisolone in IVF patients 910111213. In vitro studies have shown that aspirin attenuates placental apoptosis, and this could be a possible explanation of how aspirin is beneficial, even in the absence of endometrial or oocyte improvement 14. Proponents of aspirin consider treatment to be a simple, inexpensive, and harmless means to improve IVF outcomes 7.

However, some studies have shown anticoagulation therapy to be ineffective, and sometimes detrimental, during IVF. A large randomized controlled trial of low-dose aspirin by Urman et al found no difference in implantation or pregnancy rates in patients undergoing ICSI 4. A higher incidence of ectopic pregnancy was found in the aspirin group. A prospective, randomized, placebo-controlled IVF trial by Stern and colleagues found no benefit with aspirin and heparin for women with prior IVF implantation failure and antiphospholipid or antinuclear antibodies 15. Another small matched study of women undergoing frozen embryo transfer found an 11% pregnancy rate with aspirin compared to 33% in controls, although the results were not statistically different 16. Implantation rates were also lower with aspirin therapy, 2.9%, compared to 10.9% in untreated patients in this study. An uncontrolled study of IVF likewise found that outcomes were not improved with aspirin and heparin compared to conventional therapy 17. Finally, a prospective, randomized, double-blind, placebo-controlled trial of poor responders by Lok et al found no benefit with daily aspirin 80 mg for cancellation rates, total dose of hMG used, number of mature follicles, or number of oocytes retrieved 18. Furthermore, there was no difference in intraovarian or uterine artery pulsatility index with daily aspirin.

Randomized controlled trials have repeatedly shown that combined aspirin plus heparin improves pregnancy outcomes for women with recurrent pregnancy losses attributed to antiphospholipid antibodies 1920. This benefit is also shown in a prospective series 21. Outcomes are better with aspirin plus heparin therapy than with aspirin alone in most 2021, but not all studies 2223. Aspirin plus corticosteroid therapy, on the other hand, may be harmful. Combined low-dose aspirin plus prednisone increased the risk of preterm birth in two randomized controlled trials 2425. With a minimal benefit of aspirin alone for women with recurrent pregnancy losses and antiphospholipid antibodies, it is not surprising that we failed to find a beneficial effect of aspirin therapy in our general IVF population.

In our study, we did not test for uterine blood flow or routinely test for antiphospholipid antibodies. Therefore, we were not able to sub-divide the women in our study into groups that might be more responsive to aspirin. However, an ASRM Practice Committee Report in 1999 concluded that antiphospholipid antibodies do not affect IVF success, and therapy is not justified 26. Furthermore, we believe that implantation rates, pregnancy rates, and live birth rates are more important indicators of IVF outcome compared to indirect measurements such as endometrial blood flow. In our study, pregnancy, implantation, and live birth rates were higher in the non-aspirin control group.

Another weakness in our study is the six-year period over which our IVF cycles were reviewed. It is possible that subtle differences could bias results in the aspirin and control groups in a retrospective analysis. Additionally, the small study population yields a limited statistical power to detect minor differences in pregnancy outcomes with aspirin. There are actual and sometimes large differences between the two groups of women, which could affect the outcomes. The differencesare not significant, but might be due to the small population studied. There certainly could be minor changes in treatment protocols over that span of time, but our age-related pregnancy and live birth rates remained stable during the years of this study.

Based on the results from our study and the prospective randomized trials by Urman and colleagues 4 and Stern et al 15, aspirin is not beneficial for a general IVF population. Since implantation, pregnancy, and delivery rates are higher for non-aspirin users, our study raises the possibility that aspirin may lower IVF success. A potential fertility reducing effect of aspirin is plausible, since prostaglandins affect ovulation, fertilization, and implantation 27. Since aspirin inhibits prostaglandin synthesis, implantation could be compromised. Clearly, a larger, prospective randomized study with adequate power would be needed to determine if low-dose aspirin reduced IVF success.

There is some risk associated with aspirin therapy for infertility, although the extent of the risk for a healthy infertility population is unclear. One population based cohort study found that aspirin and nonsteroidal anti-inflammatory agents increased the risk of miscarriage, although a recent meta-analysis showed no increased risk of miscarriage with aspirin 2829. Although aspirin does not appear to alter the risk of congenital anomalies, first trimester aspirin consumption may increase the incidence of gastroschisis 30. Acetylsalicylic acid may reach the uteroplacental circulation and exert antiplatelet effects in the fetus and newborn, although the incidence of neonatal bleeding does not appear to be increased with maternal aspirin 3132. However, maternal aspirin may raise the risk of placental abruption and antenatal, intrapartum, and postpartum hemorrhage 3233. Additionally, there is at least one reported maternal death due to complications of cerebral hemorrhage in a woman treated with aspirin and heparin after IVF 34. Although these risks may be small, treatment with aspirin is not justified in the absence of a proven benefit.