When in-vitro fertilisation (IVF) began in 1978, some scientists were worried that the babies born as a result of it might suffer birth defects and health problems.
Intensive research in the early years, and a thriving population that has now grown to more than five million IVF children worldwide, have reassured scientists, but they have not stopped studying and trying to improve the process.
Recent discoveries in epigenetics - the study of how environmental factors can affect gene activity, and how a person's risk of getting chronic diseases is "programmed" into them before they are even born - have opened up new possibilities.
In IVF, eggs are removed from the ovaries and fertilised in a laboratory before being placed in the uterus. Initially, women are often given drugs to suppress their menstrual cycles before taking fertility hormones designed to make them produce more eggs to be collected and fertilised. The technique can be used to overcome infertility caused by blocked fallopian tubes, reduced fertility in the male partner or unexplained infertility, among others.
Researchers are looking into the subtle effects each part of the IVF process can have on a developing embryo - effects that continue into later life.
A priority is finding out how elements of the IVF process - such as freezing the embryos, how they are grown in the lab, how long they are kept out of the mother and how they are stored - could be responsible for those effects and how tweaks to the process could make it even safer.
Much of today's research stems from the Barker hypothesis, which proposes that birth weight may be linked to the likelihood of getting certain diseases. IVF babies are known to have lower average birth weights - even if the difference, at about 20g-30g, is small. Scientists are now investigating whether IVF conception equates to more hospital admissions, and an increased risk of cardiovascular disease, high blood pressure and diabetes in later life.
Researchers take great pains to point out that the risks for an individual born after IVF are tiny - but they could be significant for society.
"On an individual level, things that children will do in their later life will have much more influence on their health," says Professor Daniel Brison, scientific director of the department of reproductive medicine at Central Manchester and Manchester University Hospitals NHS Foundation Trust, who is looking at the long-term health of IVF children as part of the EpiHealth consortium, an EU research project.
But across the broader population, the subtle effects the process may have on those born from IVF could have an impact - for example on the kind of healthcare they might need in later life, he says.
"We have 2 per cent of babies born through IVF every year, so this is an important issue for them and for society."
Brison's team is linking a database of 110,000 IVF births from 1991 to 2009 held by the Human Fertilisation and Embryo Authority (HFEA) to other health databases to investigate patterns in IVF children's health.
His is not the first such study. Alastair Sutcliffe, professor of general paediatrics at University College London's Institute of Child Health, carried out a study last year using the HFEA database and cancer registries that showed no overall increased risk of cancer among children born from IVF compared with the population at large.
"We will revisit the data in five years to see if we continue to get good news, as cancer is a major outcome and normal risk is a measure of normal resilience," Sutcliffe says.
Brison's project is linking the HFEA babies' data to Scottish medical records to find out whether IVF babies have an increased risk of hospital admissions and growth abnormalities.
"We're interested in following up the long-term health of babies, but it's a long time to wait," Brison says.
Canadian Singer Celine Dion and husband Rene Angelil with their twins Nelson and Eddy (then 5 months old) and son Rene-Charles in 2011. All three children were conceived using IVF. They are now aged 4 and 13. Photo / Getty Images
That's where the study of early human embryos comes into its own. As part of EpiHealth, they are also being studied to determine whether the length of time they are kept in culture in the lab, or whether they are frozen, changes the embryos in a way that could make them more susceptible to disease in the long term.
Patients at Brison's Manchester fertility clinic who have used all the embryos they need to conceive their children are given the option of donating their extra frozen embryos, either to others to help them conceive, or to research such as this. "We see that human embryos have stress pathways and they do seem to be changed in response to different conditions in the environment," Brison says.
"This may allow us to improve IVF conditions to reduce stress to embryos and increase success rates and safety. IVF has been of tremendous benefit to millions of couples - we just need to make sure we're practising this technology in the safest possible way."
He is also working closely with Tom Fleming, professor of developmental biology at the University of Southampton, who is taking advantage of the shorter lifecycle of mice to gather data from in-vitro fertilisation of their embryos.
Fleming's team, also part of EpiHealth, is focusing on the effects of maternal age, of embryo vitrification - a form of freezing - and of different periods of embryo culture: three common themes within IVF-related medical treatment.
The mouse studies can be used as a starting point to look for similar effects in humans. The team has found that mouse embryos transferred to the mother later in their development as blastocysts - fertilised eggs that are left for five or six days to mature before transfer - have a greater risk of cardiovascular problems in adulthood.
Blastocyst culture is often used in human IVF because it allows embryologists to select the embryos that seem to have the best chances. The mouse study led clinicians to study data on human IVF children and find that they, too, showed signs of raised blood pressure. "There's evidence that some aspects of the cardiovascular and metabolic health of IVF children is affected - high blood pressure and body fat composition may be different, as well as evidence of more advanced bone age," Fleming says.
"There are a number of not life-threatening, but important changes in health, and the population is still young. How the health of these people will be in the future, we honestly don't know," Fleming adds. Some of these findings seem to fit in with the results of animal studies that suggest some of the technologies used to aid fertility may increase the risk of cardiometabolic diseases in later life.
Studying mice also has another advantage: they are not generally infertile, meaning that the effects of the different technologies on the embryos can more easily be seen.
Nick Macklon, professor of obstetrics and gynaecology at the University of Southampton, who is also carrying out research as part of EpiHealth, agreed that separating the effects of underlying infertility from the effects of the process to correct it was a challenge for IVF researchers, for example when looking at birth weight.
"One of the problems has been finding out if it is something about the parents' infertility itself which means that when the baby is born it's slightly lighter, or if it's the IVF process. Where the consensus seems to be going is a bit of both."
Macklon, who is also director of the Complete Fertility Centre in Southampton, foresees a move towards more natural IVF that is less reliant on administering hormones to the mother. "I think what we're seeing is a general trend towards recognition that the more natural the process can be the better."
New, safer techniques for freezing embryos open up greater possibilities for storing them until they can be transferred, so the use of hormones to artificially stimulate the mother's cycle can be avoided or minimised. Techniques to analyse the fluid in the uterus into which the embryo is placed have shown that the environment in the uterus is changed by the hormones, Macklon says. "Therefore there could be some benefit in avoiding them."
Macklon is looking for ways in which IVF patients can improve their chances by ensuring their health is as good as it can be and that their embryos are of the best possible quality.
"We have for the first time been able to analyse the nutritional content of the uterine fluid in which the embryo develops. We've found out the amino acids that are in there and we've shown that this can be affected by diet."
Fleming's team has also shown, in mice, that the mother's diet changes amino acid levels in the uterine fluid with lasting consequences.
"Poor maternal diet causes the amino acid levels to decline and embryos can sense this," Fleming says. Embryos react to the information they get from their environment, using it to "decide" how efficient the placental tissue should be later in pregnancy. "In mice, if the wrong decisions are made, it leads to long-lasting effects, associating with adult cardiovascular disease," Fleming says.
Macklon's team is halfway through a randomised study that should show whether a Mediterranean diet could help improve embryo quality.
The scientists all stress the importance of continued research into one of the most closely studied groups of people in history, who are still only in their early 20s. It will be important to study them as they reach middle age, when they are more likely to start to pick up early-onset diseases such as type 2 diabetes and cardiovascular problems, as predicted by the Barker hypothesis.
Macklon says: "We know that the environment in which we are conceived and in which we develop in the uterus can have effects right through our lives. It is important that we follow children right through into late adulthood and look at their offspring to make sure that we don't just consider the sense of reassurance we have now - which is considerable, a closed book. We'll only know the answers to some of these questions in 20, 30 years, but so far it looks good."
An ever-changing technology
Much has changed since Louise Brown, the first IVF baby, was born in Oldham, Greater Manchester, in 1978. Here are some of the key areas of innovation.
The first baby born from IVF using a frozen embryo was in Australia, in 1984. Freezing embryos maximises the chances of conception from one artificially stimulated cycle - more are produced than are needed, some can be frozen, stored and transferred at a later date. Freezing also opens up the possibility of cutting the use of artificial hormones as embryos can be placed in the mother later, at the appropriate point in her natural cycle.
The more recent introduction of a fast-freezing technique, known as vitrification, has improved the chances of frozen embryos surviving. Data from a Danish study has shown that babies born from frozen embryos are slightly heavier than other IVF babies, but it is not clear whether that is to do with the freezing process, or the fact that hormones were not needed. Professor Nick Macklon says: "We're just beginning to understand some subtle effects of freezing and thawing on the embryo. It may well be that, in the future, it will be routine to freeze all embryos and transfer them in the normal cycle."
"Using a natural cycle would be best, but it is hard," says Professor Alastair Sutcliffe. Macklon agrees - his team has been researching the hormone-stimulation phase of the IVF process.
"If you stimulate more mildly, you get better quality embryos and perhaps a healthier uterine environment. If you take that to its logical conclusion, people are now asking if one reason the babies are slightly different in weight is because of the effect of these hormones on the uterus." With advances in freezing techniques, there is a trend towards putting embryos back into a much more natural, physiological cycle, Macklon adds.
Improvements in IVF techniques and success rates have led to a trend towards transferring fewer embryos to the mother to cut risks of miscarriage and health problems related to multiple pregnancies.
A recent important development is time-lapse incubators containing cameras, allowing IVF doctors to leave embryos undisturbed between fertilisation and transfer. Previously, embryos had to be removed from their incubators every day for analysis under a microscope. Time-lapse technology also gives more detailed information on how the embryo is developing.
"This is a tremendously powerful technology which we think will improve IVF treatments considerably," says Professor Daniel Brison.
Reviewing how the embryos are cultured should be part of the drive to get as close as possible to a natural process, Macklon says.
"There are companies looking at developing devices which you can load embryos into, put them in the uterus and use the uterus as an incubator. I think that's going to be something that we'll see coming on-line in the next year or so. It's quite an important possible development."