‘Patient-friendly’ IVF

Authorities on both sides of the Atlantic have, in recent years, given birth to a concept of so-called ‘patient-friendly’ IVF (pf-IVF) [1,2]. Underlying propositions are relatively easily defined: currently prevalent IVF practices are unnecessarily cumbersome, inconvenient and costly, leading to undesirable outcomes. Logic, therefore, calls for new approaches that offer a more ‘patient-friendly’ experience. A recent opinion piece by nine highly regarded colleagues offered a good example for such thinking [3]. Claiming profound psychological distress in infertility patients as fact, they argue that current infertility treatments further enhance stress; that costs and medical procedures can lead to drop out from treatment; and that welfare and safety of patients may be adversely affected by current clinical approaches. pf-IVF, as described by Pennings and Ombelet, is proposed as obvious solution and defined as [1]:

“A policy that is cost-effective, available to the widest possible range of people and minimizes risks and burden for the patient.”

Under such a heading of ‘patient-friendliness’, they then suggest using gonadotropin-releasing hormone (GnRH) antagonists in place of agonists and, by doing so, creating a more patient-friendly IVF [3].

Although probably deserving of a review on its own, it is here not our intent to criticize the authors’ preference of antagonists over agonists; we leave this to others [4–9]. Instead, we wish to communicate that a review of the published data convinced us that this currently increasingly popular new concept, similar to other untested recent modifications to routine IVF protocols, is conceptionally flawed and has the potential of harming patient outcomes without offering compensatory benefits.

Patient entitlement to expected pregnancy rates

Since fertility treatments do not constitute medical emergencies, they offer ample opportunities to ponder treatment options. Infertility patients, therefore, are among the best educated healthcare consumers [10]. Often overlooked and/or easily dismissed, the literature convincingly demonstrates the primacy of (quick) conception over practically all other considerations for an overwhelming majority of infertility patients [10–12].

This does not mean that infertility patients are callous in regards to potential risks and/or inconveniences. They, however, knowingly, and with full understanding of risks, are choosing to make reasonable sacrifices in obtaining the primary goals of pregnancy and delivery [10,11].

Despite such obvious primacy of pregnancy success, maximal (and expected) pregnancy rates, paradoxically, are never considered part of ‘patient-friendly’ approaches towards IVF. Devroey et al. did note that [3]:

“Any protocol, intended to improve the patient experience, should not only provide advantages in terms of treatment burden, risk and distress, but also maintain IVF success.”

By emphasizing maintenance of pregnancy success, they, however, fail to explicitly endorse a concept of expected pregnancy chances, although at least recognize that patient-friendliness should not reduce pregnancy chances. They, too, thus ignore expected pregnancy rates as, likely, ultimate reflections of high patient satisfaction, while acknowledging that untested modifications to established IVF practice run the risk of actually reducing pregnancy success, as recently well demonstrated [2,13,14].

In vitro fertilization has developed into a clinically mature procedure with expected pregnancy rates, entitling patients to achieving pregnancies at those rates [15]. Compensatory benefits may, at times, have the potential to compensate for higher pregnancy rates. In absence of such proven compensatory benefits, and of informed consents permitting experimental approaches, modifications to standard practice patterns, with the potential of reducing pregnancy chances, should, however, be avoided. In other words, deviations from standard practices in IVF mandate that they be conducted only under study conditions [15].

As also its proponents accept, pf-IVF represents a major deviation from standard IVF practice. It raises the specter of potentially lowering pregnancy chances [2]. By ignoring expected pregnancy rates as a principal goal of pf-IVF, Devroey and associates, similar to others before them, therefore, incorrectly define pf-IVF [3]. Whether antagonists are more patient-friendly than agonists, is, therefore, of secondary importance. A more important question to ask would be what place do antagonists versus agonists have in maximizing IVF pregnancy chances?

The answer has remained controversial [4–9]. This is, however, exactly what makes these authors’ claim that [3] “… GnRH antagonist protocols have a number of advantages over GnRH agonists in terms of improving the patient experience, and therefore deserve at least trial use in all IVF centers” a questionable proposition.

A call for properly executed clinical trials would be acceptable, indeed, desirable. Advocating the uncontrolled introduction of antagonist protocols into all IVF centers has, however, to be viewed with suspicion since it, once again, fosters uncontrolled change in routine IVF practice under the pretence of offering patients an (unproven) compensatory benefit (i.e., alleged ‘patient-friendliness’) with potential risk of reducing pregnancy chances [7,9].

Historical examples for errors

Even if by appearance and concept sound, and offered by authorities, unproven modifications to standard IVF practice should no longer be tolerated [15–17]. Recent experiences demonstrate that even apparently obvious and logical concepts may be false; for example, nobody (these authors included) initially doubted the benefits of preimplantation genetic screening (PGS) on IVF outcomes [18]. Yet, current PGS technology was demonstrated to not achieve this goal [16,19,20] and, indeed it may, especially in older women, actually even reduce pregnancy chances [13,16].

Similar concerns also apply to the increasingly popular concept of single embryo transfer (SET) [10,15,17]. Similar to PGS, SET, by reducing twin pregnancies, presents with compelling arguments since twins carry higher maternal and perinatal outcome risks (and costs) than singleton pregnancies [21]. However, what if twins do not really denote higher risks and costs [21]?

Since most infertility patients want more than one child, correct statistics mandate outcome and cost comparisons between one twin and two singleton gestations (traditional comparisons were made between one twin and one singleton pregnancy). Under such an infertility–treatment paradigm, twin pregnancies, however, no longer demonstrate clinically relevant risk and costs increases [21].

Single embryo transfer, therefore, once again, results in decreased pregnancy chances without any compensatory benefits and leaves another, initially very logical and convincing, concept gasping for air. Strong patient desire for twin pregnancies [22], especially with advancing age and length of infertility [21,22], reduced pregnancy chances with SET [14,23–27] and increased costs (with correct assessments) [21] should make the futility of most SET apparent to even skeptics.

‘Mild’ & ‘soft’ IVF

Untested changes to routine IVF, nevertheless, continue to be proposed. Devroey et al. address so-called ‘mild’ or ‘soft’ treatment concepts, recently defined by Nargund et al. within the context of pf-IVF [28]. Similar to PGS and SET, this approach reduces pregnancy chances, especially if applied unselected. This was documented by Heijnene et al., reporting only a 43.4% cumulative pregnancy rate (leading to live births) after one full year of cycle activity. Paradoxically, they nevertheless concluded that such an approach represents cost-effective and patient-friendly medical care [29].

Most patients, of course, would consider such a cumulative pregnancy rate, in an obviously favorably selected population, anything but patient-friendly. In the USA, such a low cumulative pregnancy rate, after a full year of efforts, would, indeed, be considered unacceptable.

Kato’s group in Tokyo, which has been instrumental to introducing so-called mild or ‘natural’ stimulation, recently suggested that natural cycle IVF represents an excellent treatment option for older women [30]. The obvious question, of course, is once again what defines ‘excellent’: these investigators reported in women above the age of 38 years; depending on age, retrieval rates of 66.0–56.4%, fertilization of at least one oocyte in 77.8–76.5%, and clinical pregnancy rates per embryo transfer of 29.8–6.5%. Calculating clinical pregnancy rates per cycle start in the most favorable age group of 38–40 years, one finds, at best (assuming no cycle cancellation after fertilization, which in natural cycles is an unlikely proposition), a rate of 15.3%. In their oldest age group, 44–45 years, this rate falls to 2.8%. Both are hardly reasonable clinical pregnancy rates and do not, at least in our opinion, reflect excellent treatment options for older women with limited time to conceive.

We therefore consider the uncontrolled introduction of mild or soft treatment protocols and of other untested modifications of standard and well-established IVF procedures, outside of investigational frameworks and without use of appropriate informed consents, unethical and contrary to patient interests.

Conclusion

Patient-friendly IVF, as a universal concept, remains undefined, with a variety of unproven modification to standard, and well-established IVF protocols being arbitrarily assigned the mantle of patient-friendliness. Since practically all proposed changes in IVF practice under this heading adversely affect pregnancy chances, they create adverse outcomes, defined by lower pregnancy rates.

Any intentional reduction in pregnancy chances is ethically acceptable only if compensated for by specifically defined gains. In the absence of such documented gains, reductions in pregnancy chance have to be considered adverse outcomes, which, ethically, cannot be imposed on patients unilaterally. Since pf-IVF, in all of its proposed variations, is currently unable to define compensatory benefits for an obviously reduced pregnancy chance, all modifications to standard IVF practice summarized under this umbrella term have to be considered experimental, and should be applied only under study conditions and with appropriate informed consents.

Considering the now widely regretted premature introduction of PGS into routine IVF practice, untested, hypothetical concepts should, more than ever, be viewed with careful skepticism. Paradoxically, the opposite appears to be taking place: the recent mandate from the British Human Fertilisation and Embryology Authority (HFEA) to reduce twin pregnancies in the UK over 3 years from 24 to under 10% (under threat of licensing consequences to IVF centers) probably documents this trend best [31].

Owing to the prominence and involved authors, the recent pf-IVF manuscript by Devroey and associates also deserves special notice [3]. It is an excellent example of how even the best among us can be misled by our personal belief systems.

Premature utilization of PGS has significantly damaged IVF. Repeat performances with either SET, pf-IVF or any other untested major modification to standard IVF protocols, without appropriate statistical validation, will lead to further loss of credibility for reproductive medicine. Unchecked introduction of new treatment philosophies, including pf-IVF, even if supported by well-known and respected authorities, should, therefore, be curtailed.

Since this editorial was submitted by the authors, the Practice Committee of the American Society for Reproductive Medicine published a revised definition of what constitutes an ‘experimental procedure’. It states [32]:

“Procedures for diagnosis or treatment of infertility will be considered experimental or investigational until the published medical evidence regarding their risks, benefits, and overall safety and efficacy is sufficient to regards them as established medical practice.”

Based on this definition, pf-IVF, as suggested in this editorial, has to be considered experimental.

Financial & competing interests disclosure

The authors are supported by the Foundation for Reproductive Medicine, NY, USA, and have recieved intramural funds from the Center for Human Reproduction, NY, USA. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.