Global Nephrology: Better Cars in Japan Maybe, but also Better Dialysis?

One of the comments to my article: ”Follow-up to the 2009 Boston Conference” was as follows

“AAKP still publishes an editorial by Eli Friedman, trying to make the case that the difference in outcomes between the US and abroad is merely selection bias. One of the most outlandish claims is that the superior Japanese survival rate (the best in the world) can be entirely explained away due to much lower transplant rates in Japan, thus biasing the Japanese sample with healthier subjects who would have received transplant in the US." (By Anon, Jan 13, 2012)

Source: USRDS Atlas 2011

Friedman in his article from November 2000 suggests 3 possible explanations for the higher mortality rate in the US:

1. The United States has a higher incidence of ESRD than any other part of the world. (African Americans and Native Americans attribute to higher rates).

2. The United States treats ESRD in "marginal patients" who are otherwise quite ill. (i.e. extensive spread of cancer).

3. Other countries do not treat a substantial portion of ESRD patients who die from other listed causes.  The United States has over twice the treatment rate for ESRD than other industrialized European countries, while Canada treats approximately one-half the number of patients.

So, I decided to explore the question: Is there a plausible explanation for why there is such a staggering difference in mortality?

First, an epidemiologic update. A paper by Nakai in the February 2012 issue of Therapeutic Apheresis and Dialysis (essentially a Japanese dialysis journal) provides a pretty rich assessment of dialysis therapy. Here is the summary:

• In 2009, 290,661 patients in Japan were on dialysis. An increase of 7240 patients (2.6%) compared with that of 2008.
• The crude death rate of dialysis patients from the end of 2008 to the end of 2009 was 9.6%.
• The mean age of the new patients introduced into dialysis,  67.3 years
• The mean age of the entire Japanese dialysis patient population, 65.8 years.
• Primary diseases such as diabetic nephropathy and chronic glomerulonephritis for new dialysis patients, showed a percentage of 44.5% and 21.9%, respectively.

From the 2010 survey I learnt that:

• Among the patients treated by facility dialysis, 95.4% of patients were treated three times a week, and the average time required for one treatment was 3.92 ± 0.53 (SD) hours.
• The number of patients using a polysulfone membrane dialyzer was the largest (50.7%) and the average membrane area was 1.63 ± 0.35 m.
• Nearly 90% of patients had a functioning arteriovenous fistula and 7.1% an AV graft graft. 

Source: USRDS Atlas 2011

So how does that compare with US data?

• The US crude mortality (see Fig), while falling is two-fold higher than that in Japan (In 2006, 20.1% of U.S. dialysis patients died), but the rate is falling. 
• The higher mortality does not seem to be explained by frequency of dialysis: ≈95% of Japanese patients on three- times a week dialysis, whereas ≈98% of US patients on three times dialysis.
• The average time on dialysis of the US dialysis population is age dependent: ≈65% on >4 hrs of dialysis but a only approximately 55% of older patients on >4 hrs of dialysis.
• The age and the prevalence of diabetes was similar between the US and the Japanese.
• There is a markedly higher rate of fistula usage in Japan as compared to the US.
• The Japanese use highly purified water, but the use of the "ultrapure water" only really began when the Japanese standards were revised in 2008. ("ultrapure" dialysate is commonly defined as having a bacterial count less than 100 CFU/L and an endotoxin content less than 0.03 EU/mL measured by the Limulus amebocyte lysate assay).
• The use of home dialysis is higher in the US than Japan, although low in both. (In the US the rate of home dialysis has gone from 6% in 1985 to 0.1% in 2009).

The bottom-line? There are a number of differences in dialysis practice that alone or together could explain the difference in mortality. While the very low transplantation rate in Japan, i.e. selection bias, might be a part of the explanation, it does not explain the two-fold higher mortality rate in the US. In Japan, like the US, there is a highly prevalent diabetic population on dialysis with an average age in the 60s that similarly dialyzes 3 times a week in center. It's either better dialysis or some other factor, but not just differences in transplant rates, nor is it because the Japanese are very selective in who they treat on dialysis.