Skip Navigation


Department of Health and Human Services
Appellate Division

SUBJECT: Pancreas Transplants # 35-82

DATE: July 1, 2005




Docket No. A-04-89
NCD Ruling No. 1


Procedural history of the case

On July 8, 2004, the Board accepted a complaint filed by the aggrieved party (1) seeking review of the Medicare National Coverage Determination (NCD) addressing Pancreas Transplants in section 35-82 of the Medicare Coverage Issues Manual, as amended by CMS Transmittal Number 123 effective October 1, 2000. (2) On August 3, 2004, the Centers for Medicare & Medicaid Services (CMS) produced the administrative record relating to the NCD (NCD Record). Subsequently, CMS filed, but later withdrew, a motion to dismiss the appeal. On October 6, 2004, the aggrieved party filed a statement as to why the NCD Record is not complete or not adequate to support the validity of the NCD, as provided for in 42 C.F.R. 426.525(a). On January 7, 2005, CMS responded as provided for in 42 C.F.R. 426.525(b).

Regulatory process and standard of Board review

The applicable regulations provide that, once the aggrieved party has submitted its statement of why of an NCD is not valid and CMS has responded, the Board will evaluate whether the NCD Record is complete and adequate to support the validity of the NCD. 42 C.F.R. 426.525(c). In its review, the Board will uphold the challenged policy (or a challenged provision or provisions of a policy) if the Board finds that the relevant "findings of fact, interpretations of law, and applications of fact to law" by CMS in the NCD Record are reasonable and support any challenged provisions of the NCD in light of the relevant materials submitted to the Board by both parties. 42 C.F.R.  426.525(c)(1); 426.110 ("reasonableness standard"). Based on its evaluation, the Board may take one of two steps:

(1) If the Board finds the NCD Record to be complete and adequate to support the validity of the NCD, the Board will issue a decision to that effect, which would end the review process; or

(2) If the Board finds the NCD Record not to be complete and adequate to support the validity of the NCD, the Board will go forward with the review process by permitting discovery and the taking of evidence, including holding a hearing if appropriate.

42 C.F.R. 426.525(c)(2) and (3).

Arguments of the parties

The aggrieved party states that she needed and received a solitary pancreas transplant at the University of Pittsburgh Medical Center. She avers that the cost of that transplant was not covered by Medicare because the NCD regarding pancreas transplants (No. 35-82) permits coverage of whole organ pancreas transplantation only simultaneous with or subsequent to a kidney transplant. (3) The aggrieved party argues that the evidence cited in the NCD Record for excluding coverage of pancreas transplants alone (PTA) has been supplanted during the period after issuance of the NCD by data supporting coverage in circumstances such as hers, i.e., for a patient with hypoglycemic unawareness and other complications but without renal failure. The aggrieved party argues that more current data establishes the following:

1. PTA has become a much more common procedure since 1995 and is no longer considered experimental or investigational.

2. Patient screening practices for PTA have identified suitable candidates.

3. Private insurers now cover PTA procedures.

4. Outcomes have greatly improved for PTA recipients due to medical advances.

5. PTA graft survival rates now exceed those of simultaneous pancreas and kidney transplants (SPK) or pancreas after kidney transplants (PAK) at the time CMS approved SPK and PAK procedures for Medicare coverage.

In support of these assertions, the aggrieved party provides information on transplant frequency and graft survival rates from the Organ Procurement and Transplantation Network, information on private insurer policies, and a number of journal articles (which cite additional journal articles). She also provides a letter from her surgeon, referring to the transplantation as a "life-saving procedure, in that it eliminates hypoglycemic unawareness."

In its response, CMS describes its process for developing the NCD and the rationale for including the restriction of coverage for pancreas transplants to kidney transplant recipients. CMS argues that, even assuming the new evidence to be probative of the point that graft survival rates for PTA are now comparable to the previous survival data for SPK or PAK, that conclusion would not necessarily compel extending coverage. In discussing its basis for distinguishing between PTA and SPK and PAK, CMS refers to several issues. First, CMS points out that patients who must have kidney transplants for renal failure are already committed to an immunosuppressive regime. Therefore, CMS states, even if graft "survival rates were identical, the agency would still need to balance the added risk to the patient of immunosuppressive regimens that would not otherwise be provided to the PTA patient," including the potential for increased risks of cancer and of infections. CMS Response at 15, citing NCD Record at 127, 169. Second, CMS argues that evaluating the risks and benefits of PTA requires comparison to the results of available alternative therapy for those patients without renal failure, and that the aggrieved party had failed to explore this comparison. Id. Third, CMS argues that the NCD Record demonstrated a significant difference in the number of PTA procedures performed, especially compared to SPK procedures, which CMS interpreted as suggesting that the PTA procedure "was not as well established as SPK." Id. at 16, citing NCD Record at 39, 60-73. Fourth, CMS argues that another significant difference for solitary pancreas transplant candidates was that they would not benefit from the stricter regulatory standards governing renal transplant centers. Id. at 16, citing 42 C.F.R. 405,2170, 405.2171.


The NCD provision at issue here was adopted by CMS on the basis that there was not sufficient evidence at the time that pancreas transplants for conditions other than those identified as covered were "reasonable and necessary for the diagnosis or treatment of illness or injury" under section 1862(a)(1)(A) of the Social Security Act. 42 U.S.C. 1495y(a)(1)(A). The relevant portion of the NCD reads as follows:

Pancreas transplantation is performed to induce an insulin-independent, euglycemic state in diabetic patients. The procedure is generally limited to those patients with severe secondary complications of diabetes, including kidney failure. However, pancreas transplantation is sometimes performed on patients with labile diabetes and hypoglycemic unawareness.

Medicare has had a long-standing policy of not covering pancreas transplantation, as the safety and effectiveness of the procedure had not been demonstrated. The Office of Health Technology Assessment performed an assessment of pancreas-kidney transplantation in 1994. It found reasonable graft survival outcomes for patients receiving either simultaneous pancreas-kidney transplantation or pancreas-after-kidney transplantation.

Effective July 1, 1999, Medicare will cover whole organ pancreas transplantation . . . only when it is performed simultaneous with or after a kidney transplant . . . .

Pancreas transplantation for diabetic patients who have not experienced end stage renal failure secondary to diabetes continues to be excluded from Medicare coverage. . . . There is not sufficient evidence at this time to support a determination that these procedures are reasonable and necessary.

NCD Record at 2-3 (Transmittal 124, at 2-3).

As CMS recognizes, the aggrieved party here does not challenge the validity of this NCD when originally issued. CMS Response at 2. Yet, CMS provides little response to the core argument of the aggrieved party, i.e., that the factual premises for the exclusion of PTA are no longer reasonable, as applied to patients like the aggrieved party, because of changes in medical practice and newer scientific studies. The only direct mention CMS makes of the specific evidence proffered by the aggrieved party on this point are in two footnotes which state that this "new evidence does not have the potential to significantly affect the Board's evaluation of the NCD" and that "CMS has not evaluated this new evidence." Id. at 2, n.1, and 14, n.19. (4) In general, CMS points to material in the NCD Record which it contends supported its decision to exclude PTA procedures and then argues that, even were the data to have changed on transplant outcomes, that might not suffice to overcome the concerns identified about PTA procedures.

We first discuss briefly the points CMS relies on from the NCD Record. We then address the question of whether the considerations CMS identifies imply that new evidence offered by the aggrieved party is insufficient to undercut the reasonableness of the factual or legal underpinnings of the NCD provision excluding coverage of all PTA procedures, as applied to patients with hypoglycemic unawareness and other complications, but without renal failure.

The NCD Record includes a Decision Memorandum, dated December 16, 1994, and a Technology Assessment, dated July 1995. NCD Record at 115-79. Many of the concerns relating to pancreas transplants referenced by CMS in its brief were extensively developed in these documents. The references, however, are all dated prior to the 1995 assessment, as might be expected. Id. at 174-77. In addition, the assessment expressly noted that the request for evaluation did not include PTA, but rather only SPK and PAK. Id. at 127. Hence, it is not even clear that the assessment fully developed the evidence regarding PTA available as of the date of its issuance. Nor is it clear that the assessment considered the risks of alternative treatments for patients with hypoglycemic unawareness and other complications.

CMS may be correct that it reasonably distinguished between SPK and PAK on the one hand and PTA on the other hand, based on the NCD Record and the medical and scientific issues identified in that record regarding the risks of otherwise unneeded immunosuppression, the relative lack of information on PTA results, and the benefits of joining more novel pancreas transplant procedures with the protections of the renal transplant center provisions.

That question is, however, simply not before us. Rather, we are evaluating now whether the NCD Record can still be considered complete and adequate to support the validity of the NCD provision at issue in light of developments since 1999 (when CMS revised the NCD to permit non-Medicare covered kidney transplants, but did not have a new assessment performed). The data from the 1994-95 assessment and memorandum remain relevant nevertheless, as they establish a reasonable basis for CMS's concerns, placing a burden on the aggrieved party to demonstrate that the situation has changed to such a degree that the record can no longer be considered complete and adequate to support the validity of the NCD provision, under the reasonableness standard.

CMS does not directly deny the aggrieved party's contention that the state of practice relating to pancreas transplants has changed in the intervening years. While CMS does make a blanket assertion that the new evidence is not significant, it does not offer any analysis of the weight of that evidence or any reason to disregard it. Moreover, on its face, the proffered evidence does appear to show reduced risks from the procedure, and to indicate that some experts have determined that, properly managed, those risks are outweighed by the benefits to patients with hypoglycemic unawareness and other complications of diabetes. The aggrieved party proffers data to support her contention that changes in procedure selection criteria and methods, as well as improvements in immunosuppressive therapy, reduced the risks for PTA recipients and at least narrowed the gap identified in the NCD Record from the outcomes for SPK and PAK recipients. CMS has, as it notes, not yet considered or carefully reviewed that evidence, so that we have no prior agency determination regarding it to which we would defer if reasonable. As noted, CMS will have an opportunity later in the present proceedings to consider and respond to all new evidence, as well as to present any other evidence that might support a different conclusion.

The aggrieved party cites evidence suggesting that PTA graft survival rates as of 2003 exceeded the rates for both SPK and PAK at the time they were covered by the NCD and that PTA graft survival rates are now roughly comparable to the present rates for PAK. Aggrieved Party Statement at 3-4, att. 1, 2. CMS in its brief rejects the proposition that improved or even equivalent graft survival rates for PTA, as compared to SPK and PAK, in themselves automatically compel extension of coverage to PTA recipients. This is true (since CMS identified additional concerns as a basis for distinguishing PTA) but does not end the matter. The aggrieved party provides more than statistics on graft survival rates to support her position that the NCD exclusion of all PTA is not now reasonable.

In relation to the increased risk of immunosuppression when not otherwise needed, she proffers articles from Dr. David E. R. Sutherland (who was cited by CMS in the NCD Record) published from 1998 and 2001. Aggrieved Party Statement at 5-6, att. 6, 12. The reports, if unrebutted, show that immunosuppression therapy has improved significantly and that experts found the management of immunosuppression in appropriate PTA recipients easier than the management of their diabetes without the transplant. The aggrieved party argues that the need for immunosuppression therapy alone should not bar a transplant even when alternative treatments are available (noting that dialysis is available for renal failure, yet kidney transplants are covered in appropriate cases).

Dr. Sutherland asserted in his 2001 paper that, for nonuremic diabetic patients with hypoglycemic unawareness, the only alternative treatment is a strict insulin control strategy which leads to high glycosolated hemoglobulin levels and increased risks of secondary complications. Id., Att. 6, at 6. Hence, the aggrieved party does present information comparing PTA to alternative treatments for the type of patients she contends should be permitted the PTA alternative. Further, the aggrieved party submits evidence that, for certain patients like her, insurance companies and the American Diabetes Association determined, after evaluating more recent evidence than that on which CMS relied, that the benefits of choosing PTA may outweigh the risks, when appropriate selection criteria and other safeguards are applied.

In relation to the relative lack of experience with the PTA procedure, the aggrieved party proffers data purporting to show that the number of PTA transplants performed annually increased about five-fold since the year reviewed in the development of the NCD (1995). As with the other areas of concern raised by CMS, we are not at this stage determining that this particular increase in experience with the PTA procedures in fact occurred, nor that any increase is sufficient to make the continued non-coverage decision unreasonable. We determine only that the evidence proffered suffices to require us to go forward to take evidence from both parties.

As far as the value of regulatory controls over renal transplants, CMS does not clearly explain why this point would support continued exclusion of all PTA coverage. Since PAK transplants occur at a later date than the kidney transplants, it is not evident on this record that the pancreas transplants covered under the current NCD language would necessarily occur at a Medicare-participating renal transplant center. Furthermore, since CMS removed the requirement that any prior kidney transplant have been covered by Medicare in August 1999, it is not evident that all prior kidney transplants would have occurred in such a facility either. While the need to assure the competence of the facility performing a transplant may reasonably justify limits on Medicare coverage for PTA, it does not alone constitute a reasonable basis justifying the exclusion of all PTA coverage.

It is important to reiterate that, at this point in the proceedings, the aggrieved party need not have demonstrated that coverage is reasonable and necessary, but only that the NCD Record is not complete and adequate as a basis for the NCD provision excluding coverage. We conclude that the aggrieved party has met that burden and that therefore the proceedings should continue to the next stage.


For the reasons explained above, we conclude that the NCD Record is not complete and adequate to support the validity of the provision excluding Medicare coverage of all PTA procedures. The NCD Record to date does not suffice to sustain the validity of the challenged aspect of the NCD without taking and evaluating new evidence. Therefore, as required by the statute and regulations, we must go forward to permit discovery and to take evidence. The accompanying letter to counsel details the next steps to be taken in that process.

Our evaluation does not address the ultimate question of whether the NCD as currently written is valid under the reasonableness standard articulated in the statute and regulations.


Cecilia Sparks Ford

Donald F. Garrett

Judith A. Ballard
Presiding Board Member


1. In order to preserve the privacy of the aggrieved party, we have avoided using a name and refer only to the party status throughout this ruling.

2. CMS is converting its manuals to online versions. The current version of the Pancreas Transplant NCD is at section 260.1 of the online Medicare Coverage Issues Manual, CMS Publication 100-3.

3. A prior version of the NCD, effective July 1, 1999, required that any prior kidney transplant also have been covered by Medicare but this requirement was removed in August 1999. The NCD was revised, effective October 1, 2000, to change the applicable code, but reconfirmed that "pancreas treatment for diabetic patients who have not experienced end stage renal failure secondary to diabetes continues to be excluded from Medicare coverage." NCD Record at 3. The issuance states that "[t]here is not sufficient evidence at this time to support a determination that these procedures are reasonable and necessary." Id.

4. CMS also noted that, if the Board should disagree with its assessment of the potential for the new evidence to have a significant effect on the evaluation of the NCD's validity, the regulations at 42 C.F.R. 426.340 provide a process for handling new evidence. That process applies, however, only after the taking of evidence, not before the Board has determined whether it needs to take evidence, although nothing prevents CMS from independently reviewing the materials submitted so far and/or reconsidering the NCD at any time.