I mentioned this study once before, briefly criticizing it for (yet again) misleading the public about the mortality risk of living kidney donation. I left the rest of the paper’s inadequacies to be addressed at a later date.
Now, prodded by my living donor advocate cohort Donna Luebke (who gets a co-byline for this post), we’ll delve into some of its finer points.
First, notice the title – Long-Term Survival. By definition, ‘long-term’ refers to twenty years or more. Yet this population was only a median 6.3 years post-donation (range of 3.2-9.8 years). Hardly meets the accepted criteria for long-term.
The authors chose this sample of living kidney donors because they were pulled from OPTN’s database, which didn’t begin collecting any information on living donors prior to 1994 (even though living donors have been utilized for their kidneys since 1954). More problematic from a research standpoint is that OPTN collected ONLY social security numbers until 2000 when they began requiring transplant centers to report one year’s worth of follow-up (changed to two years in 2006). Worse, OPTN’s own data task force reported their data is “incomplete beyond the point when the discharge form is submitted (up to 6 weeks post donation) and therefore data collected beyond these time points are useless for research or making conclusions about living donor safety”*
Hmmm, that doesn’t bode well.
Two – The authors referred to their study as ‘prospective’. In research parlayance, this means that subjects are identified first and then followed forward in time. However, this study plucked its subjects from a database of people who had become living donors in the past. To truly be prospective, the researchers would’ve taken the pool of people presenting as possible living donors and followed them through their evaluations, surgery and many, many years beyond. Clearly that’s not what happened here. Same with their comparison cohort, which brings me to this convenient segueway….
Three – In order for the researchers to determine if living kidney donors die before non-donors, they must be compared to something. They can’t be thrown up against the general population becauase living donors must pass certain physiological/anatomical benchmarks before being approved as a donor. For valid research results, the comparison must be appropriate – apples to apples vs. apples to oranges, so to speak.
The OPTN database contains approximately 80,000 living kidney donors. Instead of finding 80,000 similiar non-donors, the authors pulled 9300 folks from the National Health and Nutrition Examination Survey (NHANES III)and replicated them until reaching 80,000.
If that made the warning bells go off in your head – good, you’re paying attention. Using one person 7 times is not the same as utilizing 7 different people.
And did I mention that the cohort was drawn from the NHANES III database from 1998-1994 wherein the living donors were from 1994 and beyond?
Four – The goal of this research was “to extend previous studies of live donor outcomes to a large generalizable cohort of all live kidney donors in the U. S. during a 15 year period.”
As mentioned, OPTN only began identifying living kidney donors in 1994, leaving approximately 1/3 of all living kidney donors out of this paper’s analysis. As prior researcher’s have stated, just because you know how a certain percentage are faring doesn’t mean you know anything about the others.
Also during the time period covered, surgeons moved from an open nephrectomy to a laparscopic approach. There was a ‘steep learning curve’ during this shift, with complications and mortality not previously seen. To whit:
“Although not statistically significant, it is entirely possible that the increase in mortality between 1994-1997 and 1998-2005, and the subsequent reduction thereafter, reflects the learning curve of new technology”
Remember: the subject of this paper is long-term mortality, not deaths within days of surgery or as a result of post-op complications.
Five – Not surprisingly, the researchers state “Evidnece to data suggests kidney donation is safe”. What evidence – the followup the industry refused to do on their own and has failed to do even though there’s a federal mandate? They’re referring to the U of Minnesota Ibrahim study of course. We’ll have a separate post about that, I assure you but for now, know that it only followed living donors 10 years out (not 20) and reported data on a small cohort (255 white women who donated before 40) and 40% of their total living donors declined to participate. This is not evidence of anything, despite how the mainstream media ran with it.
The researchers admit that most mortality/death estimates come from self-report and/or literature reviews. Once again, surgeons have shifted from the open to the laparscopic procedure. For validity’s sake, the authors should be separating those two populations. Unfortunately, because no one has collected accurate data on living donors, there’s no way to know which LDs had which surgery.
Finally Six (which is my favorite) – The authors state, “Regardless of what physiologic changes occur in healthy adults after kidney donation [note: do they mean drop in GFR, hyperfiltration and hypertrophy?], our findings of similar long-term survival between donors and health comparison patients suggest that these physiologic changes do not result in premature death.”
Okay, what? Premature death? They only followed these people for six years. If they were 40 when they donated, they’d be 46 now and how many healthy 40 year olds die at 46? What kind of measure is that?
But more importantly, premature death is something that can only be determined at the tail end of life. The first living kidney donor ever, Ron Herrick, should’ve lived to 85 as a white hetero male in northeast US. Instead, he died at 79 with diabetes, a stroke, end-stage renal disease, on dialysis and two months after heart surgery. Interesting that none of that made the papers.
*From an official OPTN data task force report, January 2009. Hells bells, they only just recently began verifying they have the correct social security numbers; you can imagine what that says about the rest of the data!
Segev DL, Muzaale AD, Caffo BS, Mehta SH, Singer AL, Taranto SE, McBride MA, & Montgomery RA (2010). Perioperative mortality and long-term survival following live kidney donation. JAMA : the journal of the American Medical Association, 303 (10), 959-66 PMID: 20215610