The liver transplant system tried to fix its sex bias. Did it work?

In theory, getting a donor liver in the United States when you need one should be an extremely equitable process: The sickest patients with the greatest chance to benefit from a donated organ should jump to the front of the line regardless of their gender, race, or other factors. 

But in practice, things look quite different. 

One study from 2020 showed that women were 8.6 percent more likely than men to die waiting for a liver transplant and 14.4 percent less likely to receive a deceased donor liver transplant compared to men. That gap may not seem enormous, but it still represents a significant disparity. As it turns out, it’s rooted in the very system that’s supposed to make the process fairer, a process that unintentionally ranks women lower on the waitlist for livers.

The system in question is known as the Model for End-Stage Liver Disease (MELD). The scoring model has been used since 2002 to rank candidates on the waiting list based on how likely they are to survive without a liver transplant in the short term. The higher your score, the reasoning goes, the sicker you are and the more urgently you need a liver transplant.

One of the values used to calculate the score is creatinine — a waste product made by muscles as they break down proteins or other compounds to release energy. In medicine, the creatinine levels in your blood are used to determine how well your kidneys are functioning because they process and remove creatinine. 

And this is where the system designed for equity gets a little less than equitable for some patients. 

Creatinine levels are affected by muscle mass, and women typically have lower muscle mass than men. A man and woman experiencing the same level of kidney dysfunction will therefore have different levels of creatinine in the body. The MELD model doesn’t take this difference into account — and so lower levels of creatinine in a woman’s blood can make her seem healthier than she is, bumping her farther down the liver waiting list. 

(function(){function e(){window.addEventListener(`message`,function(e){if(e.data[`datawrapper-height`]!==void 0){var t=document.querySelectorAll(`iframe`);for(var n in e.data[`datawrapper-height`])for(var r=0,i;i=t[r];r++)if(i.contentWindow===e.source){var a=e.data[`datawrapper-height`][n]+`px`;i.style.height=a}}})}e()})();

In 2023, Dr. Ray Kim, a transplant hepatologist at Mayo Clinic in Arizona who developed the original system, debuted a new model called MELD 3.0. The model not only seeks to address the sex disparity in liver access but also improve the accuracy of scores for all patients. Among other changes, it takes into account the sex of the transplant candidate and uses it to adjust the MELD score by 1.33 points for women. This is meant to address the difference in creatinine levels between men and women and provide a more accurate understanding of how sick a female patient really is.

Simple fix, right? Even though MELD 3.0 took the sex disparity in liver access seriously and developed changes to combat it, early research on its efficacy found that even this new and improved system hasn’t yet brought men and women to parity on the liver waitlist. 

And the problem may lie less with what MELD 3.0 did than with what it did not do, prompting the question: Are we focusing on the wrong problem? 

Sex matters…but so does size

Before the MELD score, liver transplants were largely conducted on a first-come, first-served basis. Those who were added to the waiting list first, received livers first. If a really sick candidate was added to the waiting list late, say because of delayed diagnosis, they would almost certainly die while waiting for a liver to become available.

The original MELD fixed that problem by creating a measure of urgency that was easy to calculate. The initial model used three main blood test values — bilirubin, creatinine, and a measure of blood clotting time — and later improvements also took into account the sodium level. 

Traditionally, the MELD model assigns a score between 6 and 40 to liver transplant candidates. MELD 3.0 adds 1.33 points to women’s scores to account for the differences in creatinine levels between the sexes. The new model also measures albumin, an additional biomarker, and adjusts for interactions between the different variables.

Sommer Gentry, a professor at NYU Grossman School of Medicine and contributor to the paper on MELD 3.0, said that using sophisticated algorithms with more biomarkers not only helps to better predict risk for women, but using a more accurate model also benefits the whole population. The researchers estimate that this fine-tuned MELD 3.0 model will reduce at least 20 waiting list deaths per year. Last year, around 800 liver transplant candidates died while waiting for a liver. 

But creatinine is only one of the two main problems that create a disparity in liver access for women. The second, Gentry says, is body size, an issue that’s not addressed by MELD 3.0. A transplanted liver has to physically fit inside the body of a recipient, and women are typically smaller than men, so a very large liver cannot be offered to a small woman. But the reverse is not necessarily true because receiving a liver that is small for the potential recipient doesn’t typically harm the candidate.    

“The smaller you are, the less likely you are to be offered something that’s actually going to work for you,” Gentry said, adding that women end up waiting longer to receive a liver that will actually fit their body. 

Part of the conversation on sex disparity in liver access is based around which of the two problems — creatinine or body-size interventions — to fix first, and whether it is possible to create equity by addressing just one of the two, Gentry said. MELD 3.0 was an attempt to remove the former barrier. 

“I think that some other papers are coming out that suggest that perhaps the other aspect of size matching is still a concern and still possibly not addressed by the resulting liver transplant system,” Gentry said. 

Why is the new model still falling short?

Gentry believes that MELD 3.0 has increased liver transplant rates for female candidates since its adoption in 2023. But early research, however, suggests that the new model falls short of fixing the problem. 

Studies done on MELD 3.0’s impact on liver transplant rates for women indicate that the model increased the number of liver transplants for female candidates — but the gains were short-lived. 

An analysis published in July of nearly 10 million daily records from over 60,000 candidates on the US liver transplant registry found that rates of completed liver transplants for women increased in the period immediately after MELD 3.0 was implemented, but the increase was temporary. Women’s estimated odds of receiving a transplant drifted back to pre-MELD 3.0 levels by the end of the study period, 15 months after the new model was implemented. Women on the waiting list and women who had received a transplant also had higher scores in the three months right after MELD 3.0 took effect, but their scores later fell back to levels comparable to those of men.

Both the successes and shortfalls of MELD 3.0 demonstrate how rethinking existing policies and standards can help reduce inequities in organ access and healthcare.

The authors think that the early surge in liver transplant rates for women was a onetime catchup. MELD 3.0 pushed a group of women who had been stuck on the list — sicker than their old scores suggested — over the priority threshold all at once, leading them to receive liver transplants. 

Notably, the one effect that did persist was that under MELD 3.0, women received more “organ offers” than before. Every time a deceased donor liver becomes available, the system generates a ranked list of the patients suitable to receive that organ. Under MELD 3.0, the number of times an available liver was offered to a female candidate increased, probably because their (more accurate) MELD 3.0 scores better reflected how sick they were. But they’re not necessarily receiving those livers, presumably because of organ size issues. 

These findings suggest that revising the MELD criteria alone may not be enough to eliminate the sex disparity, especially if the body-size disadvantage is part of the problem. 

The authors say that a patient’s sex is merely a poor proxy for the real issue — we don’t have allocation systems in place to ensure that candidates with smaller body sizes also have the same robust access to livers.

Toward liver donation equity for all

If we could figure out how to better address questions of body size in organ donation, the benefits could be enormous. According to Gentry, body-size disparity affects the organ allocation system in general. Similar issues arise with heart and lung transplants too, where the size of the organ relative to body size has to be taken into consideration during organ allocation.

She says that a size-aware organ allocation system would therefore benefit patients on the waiting list across the board. Whether an available organ is the right fit for the potential recipient should be weighed heavily in allocation algorithms, rather than just a list of ranked waitlisted candidates. 

“That should be a larger consideration in sorting the patients who will be offered a certain organ — to ask if this organ is especially suited to one kind of patient over another,” Gentry said.

Under such a system, she said, smaller livers could be offered first to the smallest candidates on a ranked list of those who are also sickest. The assumption is that a smaller candidate will have to wait longer for a viable liver that fits them, even if they have a very high MELD score. 

For such a policy to be instituted, the Organ Procurement and Transplantation Network – a nonprofit that oversees the distribution of donor organs — would have to approve it. And before such a system could be presented to the regulatory organization, more research is required.  

It is often the case in medicine that our definition of what counts as a healthy body is measured against the average white male. This too often lowers access and quality of care for both women and people of other races. For example, the standards for creatinine levels in blood reduced access to kidney transplants for Black transplant candidates for years until recent efforts were made to start correcting it. 

In that sense, both the successes and shortfalls of MELD 3.0 demonstrate how rethinking existing policies and standards can help reduce inequities in organ access and healthcare. The conversations and critiques that followed its rollout have exposed a clear gap in organ allocation strategies, one that disadvantages a large group of patients. Closing that gap, not just in liver transplantation but across the board, could bring us a step closer to equity while also making more efficient use of the organs available to save lives.

“There’s no mechanism right now, if a small liver is available, for offering it preferentially to the small women who might not be able to accept lots of other organs that are becoming available,” Gentry said. While fixing the model to rank women more accurately on the liver waiting list is a start, including body-size considerations in the organ allocation system might be the real bottleneck to address.