Wednesday, November 26, 2008

Questions of Benefits vs Risks for the UPMC Liver Transplant Program

Last week, the Wall Street Journal published an investigative report on the liver transplant program at the University of Pittsburgh Medical Center which provided a troubling view of some the issues affecting US health care.

Let me summarize the main points, as I would like to organize them, using quotes from the article, and my parenthetic comments.

Liver Transplantation Strategies at UPMC were Aggressive and Risky



Earlier this decade, UPMC made an aggressive bid to reclaim its leadership by hiring an innovative surgeon named Amadeo Marcos, who promised to double the number of liver transplants the hospital did.

Dr. Marcos delivered on his pledge. In doing so, however, he resorted to practices that some colleagues found questionable.

These practices included:

Lowering Standards for Donor Livers



To overcome a perennial shortage of organs, he used more livers from older donors.

Also,



A shortage of transplantable organs from cadavers is a perennial constraint on the number of liver transplants. Dr. Marcos overcame this in part by using organs from so-called expanded-criteria donors -- deceased people who had been older or sicker than preferred liver donors.

In the 2½ years before Dr. Marcos joined UPMC, the average age of its deceased liver donors was 41, according to UNOS. By 2003, it was 47, or nine years above the national average.

And while in 2000 and 2001, UPMC used an average of only 10 livers a year from patients older than 65, it used 45 in 2003.

Performing Transplants on Less Sick Patients (Who Are Less Likely to Benefit)



He transplanted some of these into relatively healthy patients for whom the risk-reward calculation was less certain.

In particular,



Dr. Marcos put some of these organs into patients who were in the early stages of liver disease, say Dr. Fung and Howard Doyle, who then worked in UPMC's transplant intensive-care unit. These were patients, they say, who sometimes didn't need a transplant.

'For the first time in years, we had people dying on the operating table or in the ICU,' says Dr. Doyle, now director of surgical critical care at Montefiore Medical Center in New York. At times, according to him, patients healthy enough to walk into the hospital before being transplanted died 'because they had a high-risk liver put into them.'

Data from the Scientific Registry of Transplant Recipients show that during Dr. Marcos's time at UPMC, 30 liver recipients died within two days of surgery. That was a death rate of 2.4%, versus a national average of 1.6%.

Also



Liver patients are ranked by how advanced their disease is. Based on a series of blood tests called MELD, scores range from 40 for the sickest to six for the healthiest. Most experts now believe the risks of a transplant generally outweigh the benefits for patients with MELD scores of 14 or lower.

During Dr. Marcos's nearly six years at UPMC, it performed 441 liver transplants on patients with scores of 14 or lower, according to UNOS. That was 35% of the liver transplants performed during his tenure, and compares with fewer than 7% in the 2½ years before he arrived.

Using Live Donors (Who Donate Part of Their Livers, and Also Are At Risk From the Procedure)



He used partial livers from living donors, and then understated complications from the controversial procedure.

In particular,



Dr. Marcos sharply increased the number of transplants from living donors. In these, part of the liver of a healthy person is cut off and grafted into a sick patient. If all goes well, both pieces eventually grow to normal size. The procedure is controversial because it could be risky for the otherwise healthy donor.

UPMC did 150 such surgeries while Dr. Marcos was there, according to UNOS. No donors died. However, in 69% of the cases, the recipient had a MELD score of 14 or lower -- suggesting that UPMC was putting some living donors at risk to do transplants on patients in which the risks of the operation may have outweighed the benefits.


In addition, Dr Thomas Starzl, the liver transplantation pioneer for whom the UPMC center is named,



became suspicious of the low complication rates Dr. Marcos was reporting in adult living-donor liver transplants, say people familiar with the matter. In a textbook Dr. Marcos co-wrote, he said UPMC's rate of serious complications was zero for donors and 34% among a subset of recipients.

Dr. Starzl reviewed the 121 transplants UPMC had done involving removal of the donor's right lobe, a typical procedure in adult-to-adult living-donor liver transplants. Dr. Starzl's finding, according to people with knowledge of it: Though recipients' survival rate was only slightly lower than the national average, 60% of the recipients suffered life-threatening complications, ranging from bile-duct leaks to blood-supply problems -- nearly double the rate Dr. Marcos reported.
Dr. Starzl raised his concerns with UPMC chief Mr. Romoff and other officials, including the head of the department of surgery, Timothy Billiar, say the people familiar with the situation.

A tense six-month standoff ensued. Dr. Starzl, worried that UPMC was covering the matter up, sent his findings to a medical journal, according to people familiar with the events. Dr. Billiar asked it not to publish, on the ground that Dr. Starzl hadn't obtained patient authorization to collect the data. Dr. Billiar says that Dr. Starzl's paper would have jumped the gun on a peer-reviewed internal study he had requested from another surgeon, Wallis Marsh.

UPMC and Dr. Starzl compromised: Dr. Starzl would wait for the internal study, which would be reviewed by Pierre-Alain Clavien, a Zurich surgeon who pioneered a scale to measure complications in living-donor liver transplants. UPMC's final conclusions would be published.

In January, Dr. Marsh and Dr. Clavien confirmed Dr. Starzl's finding of a 60% rate of serious complications among recipients, documents seen by The Wall Street Journal show. The review also concluded that about 10% of the living donors had suffered serious complications, belying Dr. Marcos's claim that this number was zero.

Liver Transplantation Strategies at UPMC were Expensive



Hospitals charge $400,000 to $500,000 for a liver transplant. UPMC's transplant program produced $130 million of revenue in its latest fiscal year.

There Was No Evidence that the Liver Transplantation Strategies at UPMC Provided Benefits Outweighing Their Potential Harms

The WSJ article quoted many UPMC doctors, and gave Mr Paul Wood, the UPMC Vice President for Public Relations, multiple chances to rebut the articles' findings. Mr Wood did declare, "our core mission is nothing less than providing the best and most appropriate care for patients."

However, no one quoted in the article was willing to assert, much less justify, a claim that the aggressive tactics used during Dr Marcos' time at UPMC provided benefits that outweighed their risks, including in some cases their risks to donors as well as patients.

Another news article on the UPMC controversy published by the Pittsburgh Tribune-Review quoted Dr James Trotter, an investigator in the Adult-to-Adult Living Donor Living Transplant Cohort, did state "there's a survival advantage for people who undergo live-donor liver transplant versus waiting on the transplant list."

A quick look at the literature revealed an article from the UPMC program reporting a case-series of patients who received transplants from living donors.(1) The article showed that their survival rate was comparable to that for patients who received cadaver transplants. However, it did not include a control group, and therefore could nor provide even weak evidence that outcomes for patients receiving transplants from living donors are superior to those receiving them from cadavers. In addition, there is evidence that the risks to living donors are clinically significant. For example, a systematic review suggested that donor mortality may be more than 0.2%, and donor morbidity ranged from 0% to 90% in several studies, with a median of 16%.(2)

Thus, no defender of the strategies previously used at UPMC, nor any clinical research that I found provided strong evidence that the benefits of the use of living donors outweighed their risks to the donors, or to the recipients. Furthermore, no one put forward any evidence that the aggressive approach pushed by Dr Marcos generally provided benefits to patients that either exceeded those of a more conservative approach, or outweighed its apparent risks.

Perverse Incentives Favored an Aggressive, Risky Approach

As I noted above, the UPMC liver transplant program brought in a large amount of money, especially considering that the program provided transplants to nver more than 300 patients a year, according to a chart provided in the WSJ article. This enabled Dr Marcos to make a generous salary and maintain a lavish lifestyle.


UPMC set out to hire a surgeon who could restore the program to its former glory. It settled on Dr. Marcos, a dashing Venezuelan with a taste for Ferraris and Porsches, who specialized in the emerging field of transplants from living donors.
Furthermore,


UPMC offered Dr. Marcos $500,000 a year and "additional incentive payments," a letter dated June 21, 2002, shows. Dr. Marcos came aboard as director of clinical transplantation....

This was in a context of a medical center run by businesspeople instead of clinicians, who were increasingly richly rewarded, a context that should be familiar to Health Care Renewal readers.


UPMC is a nonprofit hospital system whose income is largely exempt from taxes. Yet, it is increasingly run like a for-profit company, paying its executives high salaries, jumping into new activities and expanding abroad. Its quest to ramp up its transplant business shows how a drive for higher revenue, now common at nonprofit hospitals, could risk compromising patient care.
Also,


Dr. Marcos's nearly six years at UPMC coincided with rapid growth at the medical center. UPMC is one of the nation's most financially successful nonprofit hospital systems, with operations ranging from Pennsylvania to Ireland and Qatar. Even though three-quarters of its $7 billion in annual revenue is exempt from federal and local taxes, UPMC has acquired many of the trappings of large, for-profit corporations.

Its chief executive, Jeffrey Romoff, earned $4 million in the fiscal year ended June 30, 2007, and 13 other employees earned in the roughly $1 million to $2 million range. For their transportation, UPMC leases a corporate jet. Earlier this year, UPMC relocated its headquarters into Pittsburgh's tallest skyscraper, the 62-story U.S. Steel Tower.
Interval Summary

So, in my humble opinion, the UPMC liver transplant story illustrates how in the US we richly reward aggressive, high-technology, cutting edge and risky care, even in the absence of any good evidence that such care provides benefits to patients that outweigh its risks. Thus, is it any surprise that we provide a lot of expensive, risky care, but may have little to show for it?

Furthermore, breaking up the medical "guild" and handing control of health care over to bureaucrats, managers, and executives paved the way for a business culture of health care that richly rewards those on top, and enables cults of personality and imperial CEOs. This has lead to a culture that puts the self-interest of leaders over the mission to care for patients. The leader-centric culture also has proved intolerant of any criticism leveled at the fearless leaders. This leads me to my last point.

A Failed Attempt to Manage the Message

We blogged earlier about how the leaders of health care organizations try to "manage the message" so as to glorify themselves and hide their own flaws. In the coverage of the UPMC liver transplant service there was plenty of attempted message managing.

First, note that no one in the UPMC leadership who bore responsibility for the transplant program was willing to say anything. Dr Marcos appeared to be in an undisclosed location.


Dr. Marcos, 46 years old when he left UPMC, did not respond to numerous attempts to reach him, including a letter sent to his home. A lawyer who represented him in a court case last year said he hadn't been in contact with Dr. Marcos for months.

Justifications of the lavish compensation given to the CEOs of health care and other large organizations often includes their heavy responsibilities. Yet the CEO of UPMC was also unavailable for comment.


UPMC declined to make Mr. Romoff available for an interview.

Instead, it was left to Mr Wood to defend the UPMC program, not only in the WSJ article, but now in several letters to the editor. Unfortunately for him, he had little ammunition for this cause.

One example was how he had to respond to the issue of whether UPMC performed transplants on patients whose liver problems were not severe enough to make them the most appropriate transplant candidates. Note that the WSJ article stated that 35% of transplant recipients during the time of Dr Marcos' leadership had MELD scores less than 15, indicating that they had favorable prognoses (without transplant.) Mr Wood responded in the WSJ article,


it wasn't until 2006 that the transplant community coalesced around a score of 15 as a cutoff to allocate organs. 'It would be unrealistic to expect a physician to practice according to yet-to-be-discovered criteria,' he said.
Actually, prioritizing liver transplants according to the recipients' MELD scores became the policy of the Organ Procurement and Transplant Network (OPTN) of the United Network for Organ Sharing (UNOS) in 2002. An article documenting the network's experience during the first year of this policy showed that the proportion of patients with mean MELD scores less than 15 ranged from about 5% to 25% across the 11 OPTN regions, with an average of about 15%.(3) This suggested there was a clear consensus to minimize the number of transplants for patients in this most favorable prognosis group by 2002. Sorry, Mr Wood.

Furthermore, yesterday Mr Wood published a letter in the Wall Street Journal in response to its article. In it, he did not provide any specific justification for the aggressive transplant policies that was based on evidence of benefits from these policies. At best, he could say,


As the pioneer and acknowledged leader in transplantation surgery, UPMC is continually exploring and testing new life-saving procedures. The use of expanded criteria organs is an attempt to alleviate a critical nationwide shortage. This alone was a compelling reason for developing a potentially groundbreaking program in living-donor liver transplant, but clearly there are risks involved whenever new procedures are developed.


This completely begged the question of whether the benefits were sufficient to offset the risks.

Furthermore, he noted,


To suggest that medical decisions are made on the basis of revenues and profits is simply ludicrous. Transplantation procedures account for less than 2% of UPMC's $7 billion in annual revenues, a fact the reporter mostly chose to ignore.
One would think an organization boasting "50,000 employees ... [and which] comprises 20 tertiary, specialty, and community hospitals, 400 outpatient sites and doctors’ offices" would pay attention to a mere 200 patients a year which provide 2% of its revenue.

He had to end up with this bit of puffery,


The real story at UPMC -- one that would clearly be of interest and importance to your readers -- is how a small, regional psychiatric hospital in Western Pennsylvania transformed itself into a self-sustaining global health enterprise that provides the highest quality medical care to patients around the world.


The advent of the internet age makes it much easier to see through attempts by health care leaders to "manage the message" (in this case, and usually by proxy) to insulate them from criticism for their actions. Maybe in the near future we can dream of health care in which incentives are made proportional to effort, ability, and most importantly the ratio of benefits to harms provided to patients.

ADDENDUM (26 November, 2008) - We were first alerted by issues with the UPMC liver transplant program in a comment on this Health Care Renewal post.

References

1. Taioli E, Marsh W. Epidemiological study of survival after liver transplant from a living donor. Transpl Int 2008: 21(10):942-7.

2. Middleton PF, Duffield M, Lynch SV, Padbury RT, House T, Stanton P, et al. Living donor liver transplantation--adult donor outcomes: a systematic review. Liver Transpl 2006; 12(1):24-30.

3. Freeman RB, Wiesner RH, Edwards E, Harper A, Merion R, Wolfe R et al. Results of the first year of the new liver allocation plan. Liver Transpl 2004; 10: 7-15.