Development of Assisted Reproductive Technology innovations: the case of Time-Lapse Monitoring

This thesis examines the development and evaluation of time-lapse monitoring, a technological innovation in assisted reproduction. IVF laboratories increasingly offer additional treatment options to couples. These ‘add-ons’ are often claimed to increase pregnancy chances despite limited high-quality evidence for their added value. Time-lapse monitoring is an innovation that has been subject to considerable hype over the past decade and is proposed to improve embryo culture and selection. Conventional IVF involves removing embryos from the incubator for regular morphological assessments, which can cause disturbances of the culture conditions. Time-lapse incubators continuously record embryo development using built-in cameras, offering uninterrupted embryo culture. These systems also generate detailed data on embryo development that can be analyzed using software or artificial intelligence to support embryo selection. Initial evaluation in a prospective, non-randomized pilot study compared embryo selection using a time-lapse system combined with morphology to conventional embryo selection. Clinical outcomes were similar between groups, despite fewer top-quality embryos being selected in the time-lapse group. These findings prompted further investigation in a large, multicenter, randomized controlled trial involving 1731 patients undergoing IVF or ICSI. Three groups were compared: uninterrupted culture with time-lapse-based embryo selection, uninterrupted culture with conventional embryo selection, and interrupted culture with conventional embryo selection. Approximately half of the patients achieved a live birth within one year across all groups. No differences were observed in pregnancy rates after the first embryo transfer or in the time to achieve pregnancy. Contrary to earlier, retrospective findings, these results indicate that neither uninterrupted culture in a time-lapse incubator nor time-lapse-based embryo selection improves clinical outcomes. Rates of serious pregnancy complications and birth weight did not differ between time-lapse and conventional methods, indicating that time-lapse monitoring is safe for patients and offspring. Cost-effectiveness analyses are often overlooked when evaluating innovations in assisted reproduction. After the publication of several large randomized controlled trials, the fertility industry and IVF clinics are now faced with the inconvenient reality that the use of expensive time-lapse incubators, even combined with machine learning or deep learning for embryo selection, results in comparable clinical outcomes to conventional methods. The affordability of healthcare in general, and of IVF treatments in particular, may be compromised when limited resources are invested in technologies that are not cost-effective. Our cost-effectiveness analysis demonstrates that time-lapse monitoring is unlikely to be cost-effective compared to conventional methods. However, time-lapse incubators may offer advantages for IVF laboratories that are not fully captured by traditional clinical outcome measures. The findings from our randomized controlled trial warranted further investigation into the effectiveness of time-lapse benchtop incubators compared to conventional big box incubators. Although differences in laboratory parameters were observed, pregnancy rates after the first embryo transfer did not improve when using time-lapse incubators. Well-designed studies incorporating cumulative outcomes and dedicated cost-effectiveness analyses are needed to further investigate the influence of incubator type on embryo development and clinical results. Overall, this thesis shows that, contrary to the expectations of many IVF professionals worldwide, the use of time-lapse incubators does not improve clinical outcomes compared with conventional methods. Our research highlights a recurring pattern in reproductive medicine: new technologies are often adopted into clinical practice before robust evidence of effectiveness and cost-effectiveness becomes available. A critical assessment of new technologies is essential to ensure responsible innovation. Innovations such as time-lapse monitoring should therefore be evaluated based on four key objectives: improving clinical results, verifying safety, reducing costs, and optimizing workflow. Healthcare professionals must consider not only which technologies to adopt, but also for whom, in order to prevent an unsustainable escalation of costs and resource consumption in the future.