New Heart Allocation Algorithm Appears Effective

SAN DIEGO – A new allocation algorithm that is designed to improve regional sharing of donor hearts with sicker patients before they are allocated locally to less-sick patients appears to be having the intended effects, according to a national cohort study.

In the study of nearly 12,000 adult patients who were wait-listed for primary heart transplantation in 2004-2009 in the United States, those who were wait-listed after the new algorithm was implemented were 17% less likely to die on the waiting list or to become too sick for transplantation, researchers reported at the annual meeting of the International Society for Heart and Lung Transplantation.

Moreover, this benefit was achieved without any increase in the rate of in-hospital mortality among transplant recipients, even though they were sicker on average.

"The risk of dying on the heart transplant [waiting list] or becoming too sick for transplant has declined [in the United States] since the change in allocation algorithm in 2006," said lead investigator Dr. Tajinder P. Singh, a pediatric cardiologist at Children’s Hospital Boston. And reassuringly, "the shift in hearts to sicker transplant candidates has not resulted in higher early posttransplant mortality."

These findings suggest that the new algorithm has been effective "not only from a utilitarian view, which means most benefit for most people, but even from the fairness or justice perspective," he commented. "By granting the hearts to sicker people, you are taking care of that point of view, too."

An attendee asked whether patterns might differ at the local or regional level vs. the national level, given that some centers in the New York City area, for example, feel they have been hurt by the new algorithm. Dr. Singh replied that because of small patient numbers and regional variations, it was not possible to get a reliable picture at those levels.

"The demand for donor hearts continues to exceed their supply," he said, giving background to the study. "The United Network for Organ Sharing has periodically modified the allocation algorithm in the United States" to improve waiting list outcomes.

The last such modification, implemented in July 2006, expanded the sharing of these scarce organs across a geographic region, making them available first to the sickest patients (those with status 1A or 1B) in a region before allocating them locally to less-sick patients.

"The goal of such a change was to lower national [waiting list] mortality without a concurrent increase in posttransplant mortality, and that consideration is more than theoretical because sicker patients will be at higher risk of dying post transplant," he explained. "The early outcome trends after the allocation change have been supportive, but regional analyses have questioned the merits of the new allocation."

The investigators studied all patients aged 18 years or older who were placed on the waiting list for primary heart transplantation between July 1, 2004, and June 30, 2009, and who were undergoing transplantation of only a heart.

For comparison, the patients were split according to when they were listed into "era 1" (before the date of implementation of the new algorithm) and "era 2" (after that date). Study results were based on 11,864 patients in total; 38% were listed in era 1 and 62% were listed in era 2.

Patients in the two eras were similar with respect to most sociodemographic and medical factors, except that those in era 2 were more likely to be aged 60 years or older (32% vs. 28%), to receive mechanical support (14% vs. 13%), and to be sicker, as indicated by having a transplantation status of 1A (20% vs. 19%) or 1B (38% vs. 32%), for instance.

Overall, 13% of the patients studied either died or had a worsening of their condition that prevented transplantation while they were on the waiting list, the study’s primary end point, Dr. Singh reported.

Before statistical adjustment, patients in era 2 were 14% less likely than their counterparts in era 1 to die or worsen while on the wait list (hazard ratio, 0.86; P = .005). And this benefit was evident among both status 1A patients and status 1B patients individually.

After adjustment for numerous potential confounders, patients in era 2 were 17% less likely to die or worsen while on the wait list (HR, 0.83; P = .001). The benefit was similar in most subgroups, except that by race, it was mainly limited to white patients.

Other risk-reducing factors included having an implantable cardioverter defibrillator (HR, 0.87) and having a continuous-flow left ventricular assist device (HR, 0.56).

Overall, 65% of the patients ultimately underwent transplantation. Compared with their counterparts in era 1, era 2 transplant recipients had a shorter median wait time before receiving a heart (55 vs. 63 days; P less than .001) and were more likely to be status 1A at transplantation (48% vs. 37%; P less than .001).

The donor ischemic time was longer for recipients in era 2 (3.3 vs. 3.2 hours; P = .02), but the small difference was probably not clinically important, according to Dr. Singh.

The lack of a greater difference in ischemic time – despite the sharing of organs over larger geographic areas in the latter era – was not surprising, he said. "The way it occurred, it went from local to within 500 miles, say. It may be broader regional sharing, but it’s not long distance to get to [the heart] and bring the heart in to the surgery."

There was no rise in the rate of in-hospital mortality post transplantation with implementation of the new algorithm. In fact, "interestingly, in-hospital mortality was lower rather than higher [in era 2], even though sicker patients were getting transplanted," Dr. Singh commented, with a rate of 5.3% in era 2, compared with 6.3% in era 1.

Dr. Singh reported having no conflicts of interest related to the research.

SAN DIEGO – A new allocation algorithm that is designed to improve regional sharing of donor hearts with sicker patients before they are allocated locally to less-sick patients appears to be having the intended effects, according to a national cohort study.

In the study of nearly 12,000 adult patients who were wait-listed for primary heart transplantation in 2004-2009 in the United States, those who were wait-listed after the new algorithm was implemented were 17% less likely to die on the waiting list or to become too sick for transplantation, researchers reported at the annual meeting of the International Society for Heart and Lung Transplantation.

Moreover, this benefit was achieved without any increase in the rate of in-hospital mortality among transplant recipients, even though they were sicker on average.

"The risk of dying on the heart transplant [waiting list] or becoming too sick for transplant has declined [in the United States] since the change in allocation algorithm in 2006," said lead investigator Dr. Tajinder P. Singh, a pediatric cardiologist at Children’s Hospital Boston. And reassuringly, "the shift in hearts to sicker transplant candidates has not resulted in higher early posttransplant mortality."

These findings suggest that the new algorithm has been effective "not only from a utilitarian view, which means most benefit for most people, but even from the fairness or justice perspective," he commented. "By granting the hearts to sicker people, you are taking care of that point of view, too."

An attendee asked whether patterns might differ at the local or regional level vs. the national level, given that some centers in the New York City area, for example, feel they have been hurt by the new algorithm. Dr. Singh replied that because of small patient numbers and regional variations, it was not possible to get a reliable picture at those levels.

"The demand for donor hearts continues to exceed their supply," he said, giving background to the study. "The United Network for Organ Sharing has periodically modified the allocation algorithm in the United States" to improve waiting list outcomes.

The last such modification, implemented in July 2006, expanded the sharing of these scarce organs across a geographic region, making them available first to the sickest patients (those with status 1A or 1B) in a region before allocating them locally to less-sick patients.

"The goal of such a change was to lower national [waiting list] mortality without a concurrent increase in posttransplant mortality, and that consideration is more than theoretical because sicker patients will be at higher risk of dying post transplant," he explained. "The early outcome trends after the allocation change have been supportive, but regional analyses have questioned the merits of the new allocation."

The investigators studied all patients aged 18 years or older who were placed on the waiting list for primary heart transplantation between July 1, 2004, and June 30, 2009, and who were undergoing transplantation of only a heart.

For comparison, the patients were split according to when they were listed into "era 1" (before the date of implementation of the new algorithm) and "era 2" (after that date). Study results were based on 11,864 patients in total; 38% were listed in era 1 and 62% were listed in era 2.

Patients in the two eras were similar with respect to most sociodemographic and medical factors, except that those in era 2 were more likely to be aged 60 years or older (32% vs. 28%), to receive mechanical support (14% vs. 13%), and to be sicker, as indicated by having a transplantation status of 1A (20% vs. 19%) or 1B (38% vs. 32%), for instance.

Overall, 13% of the patients studied either died or had a worsening of their condition that prevented transplantation while they were on the waiting list, the study’s primary end point, Dr. Singh reported.

Before statistical adjustment, patients in era 2 were 14% less likely than their counterparts in era 1 to die or worsen while on the wait list (hazard ratio, 0.86; P = .005). And this benefit was evident among both status 1A patients and status 1B patients individually.

After adjustment for numerous potential confounders, patients in era 2 were 17% less likely to die or worsen while on the wait list (HR, 0.83; P = .001). The benefit was similar in most subgroups, except that by race, it was mainly limited to white patients.

Other risk-reducing factors included having an implantable cardioverter defibrillator (HR, 0.87) and having a continuous-flow left ventricular assist device (HR, 0.56).

Overall, 65% of the patients ultimately underwent transplantation. Compared with their counterparts in era 1, era 2 transplant recipients had a shorter median wait time before receiving a heart (55 vs. 63 days; P less than .001) and were more likely to be status 1A at transplantation (48% vs. 37%; P less than .001).

The donor ischemic time was longer for recipients in era 2 (3.3 vs. 3.2 hours; P = .02), but the small difference was probably not clinically important, according to Dr. Singh.

The lack of a greater difference in ischemic time – despite the sharing of organs over larger geographic areas in the latter era – was not surprising, he said. "The way it occurred, it went from local to within 500 miles, say. It may be broader regional sharing, but it’s not long distance to get to [the heart] and bring the heart in to the surgery."

There was no rise in the rate of in-hospital mortality post transplantation with implementation of the new algorithm. In fact, "interestingly, in-hospital mortality was lower rather than higher [in era 2], even though sicker patients were getting transplanted," Dr. Singh commented, with a rate of 5.3% in era 2, compared with 6.3% in era 1.

Dr. Singh reported having no conflicts of interest related to the research.

SAN DIEGO – A new allocation algorithm that is designed to improve regional sharing of donor hearts with sicker patients before they are allocated locally to less-sick patients appears to be having the intended effects, according to a national cohort study.

In the study of nearly 12,000 adult patients who were wait-listed for primary heart transplantation in 2004-2009 in the United States, those who were wait-listed after the new algorithm was implemented were 17% less likely to die on the waiting list or to become too sick for transplantation, researchers reported at the annual meeting of the International Society for Heart and Lung Transplantation.

Moreover, this benefit was achieved without any increase in the rate of in-hospital mortality among transplant recipients, even though they were sicker on average.

"The risk of dying on the heart transplant [waiting list] or becoming too sick for transplant has declined [in the United States] since the change in allocation algorithm in 2006," said lead investigator Dr. Tajinder P. Singh, a pediatric cardiologist at Children’s Hospital Boston. And reassuringly, "the shift in hearts to sicker transplant candidates has not resulted in higher early posttransplant mortality."

These findings suggest that the new algorithm has been effective "not only from a utilitarian view, which means most benefit for most people, but even from the fairness or justice perspective," he commented. "By granting the hearts to sicker people, you are taking care of that point of view, too."

An attendee asked whether patterns might differ at the local or regional level vs. the national level, given that some centers in the New York City area, for example, feel they have been hurt by the new algorithm. Dr. Singh replied that because of small patient numbers and regional variations, it was not possible to get a reliable picture at those levels.

"The demand for donor hearts continues to exceed their supply," he said, giving background to the study. "The United Network for Organ Sharing has periodically modified the allocation algorithm in the United States" to improve waiting list outcomes.

The last such modification, implemented in July 2006, expanded the sharing of these scarce organs across a geographic region, making them available first to the sickest patients (those with status 1A or 1B) in a region before allocating them locally to less-sick patients.

"The goal of such a change was to lower national [waiting list] mortality without a concurrent increase in posttransplant mortality, and that consideration is more than theoretical because sicker patients will be at higher risk of dying post transplant," he explained. "The early outcome trends after the allocation change have been supportive, but regional analyses have questioned the merits of the new allocation."

The investigators studied all patients aged 18 years or older who were placed on the waiting list for primary heart transplantation between July 1, 2004, and June 30, 2009, and who were undergoing transplantation of only a heart.

For comparison, the patients were split according to when they were listed into "era 1" (before the date of implementation of the new algorithm) and "era 2" (after that date). Study results were based on 11,864 patients in total; 38% were listed in era 1 and 62% were listed in era 2.

Patients in the two eras were similar with respect to most sociodemographic and medical factors, except that those in era 2 were more likely to be aged 60 years or older (32% vs. 28%), to receive mechanical support (14% vs. 13%), and to be sicker, as indicated by having a transplantation status of 1A (20% vs. 19%) or 1B (38% vs. 32%), for instance.

Overall, 13% of the patients studied either died or had a worsening of their condition that prevented transplantation while they were on the waiting list, the study’s primary end point, Dr. Singh reported.

Before statistical adjustment, patients in era 2 were 14% less likely than their counterparts in era 1 to die or worsen while on the wait list (hazard ratio, 0.86; P = .005). And this benefit was evident among both status 1A patients and status 1B patients individually.

After adjustment for numerous potential confounders, patients in era 2 were 17% less likely to die or worsen while on the wait list (HR, 0.83; P = .001). The benefit was similar in most subgroups, except that by race, it was mainly limited to white patients.

Other risk-reducing factors included having an implantable cardioverter defibrillator (HR, 0.87) and having a continuous-flow left ventricular assist device (HR, 0.56).

Overall, 65% of the patients ultimately underwent transplantation. Compared with their counterparts in era 1, era 2 transplant recipients had a shorter median wait time before receiving a heart (55 vs. 63 days; P less than .001) and were more likely to be status 1A at transplantation (48% vs. 37%; P less than .001).

The donor ischemic time was longer for recipients in era 2 (3.3 vs. 3.2 hours; P = .02), but the small difference was probably not clinically important, according to Dr. Singh.

The lack of a greater difference in ischemic time – despite the sharing of organs over larger geographic areas in the latter era – was not surprising, he said. "The way it occurred, it went from local to within 500 miles, say. It may be broader regional sharing, but it’s not long distance to get to [the heart] and bring the heart in to the surgery."

There was no rise in the rate of in-hospital mortality post transplantation with implementation of the new algorithm. In fact, "interestingly, in-hospital mortality was lower rather than higher [in era 2], even though sicker patients were getting transplanted," Dr. Singh commented, with a rate of 5.3% in era 2, compared with 6.3% in era 1.

Dr. Singh reported having no conflicts of interest related to the research.