August 15, 2001 - ASM Statement on New Clinical Laboratory Tests--Payment Determinations for Calendar Year 2002

The American Society for Microbiology (ASM) appreciates the opportunity to provide comments to the Centers for Medicare and Medicaid Services (CMS) on reimbursement for new Current Procedural Terminology (CPT) codes announced for 2002. ASM is the largest educational, professional, and scientific society dedicated to the advancement of the microbiological sciences and their application for the common good. The Society represents more than 42,000 microbiologists, including scientists and science administrators working in a variety of areas, including biomedical, environmental, and clinical microbiology.

Many of our members have primary involvement in the clinical laboratory environment including individuals directing clinical microbiology or immunology laboratories, individuals licensed or accredited to perform such testing, industry representatives marketing products for use, and researchers involved in evaluating the performance of new technologies. Thus, our Society has a significant interest in the process of establishing reasonable reimbursement for medically necessary laboratory testing to ensure quality patient care for Medicare beneficiaries.

The ASM's Public and Scientific Affairs Board Committee on Professional Affairs reviewed the June 29, 2001 Federal Register Notice [HCFA-1186-N], Medicare Program: Public Meeting for New Clinical Laboratory Tests-Payment Determinations for Calendar Year 2002. Based on the payment methodology of "cross walking" and "gap filling" as defined in this notice, ASM offers the following comments with respect to reimbursement for new CPT codes announced for 2002:

Code 871XX

Cytomegalovirus (CMV) direct fluorescent antibody (DFA)

Nature of the test method:

The code for CMV DFA logically belongs in the Microbiology 872XX series of codes "intended for primary source only." For confirmation of an isolate in cell culture, code 87253 would apply. This series has a common descriptor, "infectious agent antigen detection by immunofluorescent technique; (specific agent)." The technique involves several steps as follows:

  1. Specimen preparation followed by application and fixation of cellular material to a microscope slide. In preparation of samples to assess for the presence of CMV antigens in circulating leukocytes, a time-sensitive and labor-consuming concentration step is required which justifies the additional use of code 87015 (concentration, any type, for infectious agents).
  2. Application of fluorescein conjugated antibody reagents to the prepared smear. The antibody reagents may be monoclonal or polyclonal, and products may actually include a "cocktail" of antibodies designed to optimize sensitivity in detection of a single microbial analyte. Formats may be single step (termed direct), or involve a two step "sandwich" stain process (termed indirect), but all formats are adequately covered by the proposed CPT code.
  3. Washing steps to remove excess reagents followed by mounting for microscopic evaluation. The actual number of steps varies with specific approach.
  4. Fluorescent microscopic observation to detect the presence of CMV antigens in human cells. This step is a labor-intensive and manual process that requires a registered or licensed clinical laboratory scientist who is trained to interpret patterns of specific staining compatible with CMV infection and differentiate this from nonspecific staining which is invariably present to some degree.
  5. Reporting of a qualitative, semi-quantitative, or for some applications, quantitative result. Reports also include interpretive comments about the type of staining observed and its possible clinical relevance.


The most common application of this test procedure is in the evaluation of immunocompromised patients for the presence of circulating CMV antigen, particularly hematopoietic stem cell transplant and solid organ transplant patients. This application is commonly called a "CMV antigenemia" test. This test detects a specific antigen called pp65 and quantitates the amount of antigen present. The test is used to assess risk of acquisition or worsening of CMV disease. Clinically, the results are used to determine the need to institute preemptive therapy, and for the assessment of response to antiviral therapy. This test may also be used in HIV infected patients in whom antiviral therapy is being contemplated.

The proposed CMV DFA CPT code is appropriate to identify those test formats using immunofluorescent antibody staining. However, it should be recognized that there is a comparable product available which uses an immunperoxidase-labeled antibody. This test differs from an immunofluorescence test in that the detection of specific CMV antigen can be accomplished using light microscopy. There is not currently an appropriate Microbiology infectious agent antigen detection code for this alternative but equivalent test. CPT-4 code 87332, "infectious agent antigen test by enzyme immunoassay, qualitative or semi-quantitative, multiple step method; cytomegalovirus" may be used, but the microscopic "antigenemia" test is quite different from the automated or semi-automated micro-device format. However, the National Limitation Amounts (NLA) for the DFA's and multistep EIA's are currently identical.

CMV DFA tests may also be used in the evaluation of other specimen types for evidence of CMV infection. Reagents are available from a number of diagnostic companies, but users are advised to remain current on FDA clearance status for these applications, or if using as an Analyte Specific Reagent (ASR), to use in accordance with all applicable ASR rules prior to billing Medicare.


Direct costs of performing proposed code 872XX are generally equivalent to the costs of performing any test in the "Infectious agent antigen detection by immunofluorescent technique" specific analyte series (87260-87299). The NLA for all of these codes is currently uniformly set at $16.58. However, there are additional costs for the specimen preparation component of the "antigenemia" application which justify the additional use of code 87015, concentration (any type) for infectious agents (NLA $9.23) with appropriate composite billing.

Recommendations for reimbursement determination

The addition of this code is a continuation of a trend initiated in the 1998 CPT Manual toward defining analyte specific codes for unique non-culture dependent methods for specific infectious agent detection. This code does not represent a new technology, and we believe it can be legitimately cross walked to the other single "infectious agent antigen detection by immunofluorescent technique" codes and that reimbursement should be set at 100% of the NLA for these codes. The immunoperoxidase modification of the "antigenemia" test should also be considered equivalent to this proposed code and consideration should be given to establishing a unique code for the immunoperoxidase test format.

Code 871XX

Enterovirus Direct Fluorescent Antibody (DFA)

Nature of the test method

The code for Enterovirus DFA, like CMV DFA, logically belongs in the Microbiology 872XX series of codes for "infectious agent antigen detection by immunofluorescent technique" which are "intended for primary source only." Again, for confirmation of an isolate in cell culture, code 87253 would apply. The detailed method description for an "immunofluorescent antigen" detection previously given for CMV DFA would also apply for enterovirus DFA.


Clinical applications for a direct specimen Enterovirus DFA are not well established. Further, we are unaware of any products cleared by the Food and Drug Administration (FDA) for this procedure. Any clinical use of this code would therefore likely be either investigational or in accordance with ASR rules. The primary potential use of this code would be for diagnosis of an enteroviral infection other than meningitis if use of a specific antiviral is being considered. For meningitis, testing is typically performed on cerebrospinal fluid specimens by either cell culture (87252) or a qualitative amplified nucleic acid test (87798; generic method code).


Direct costs of performing proposed code 872XX should be equivalent to the costs of performing any test in the "infectious agent detection by immunofluorescent technique" specific analyte series (87260-87299). As previously stated, the NLA is currently uniformly set for these codes.

Recommendations for reimbursement determination

The addition of this code also appears to be a continuation of the trend initiated in 1998 toward establishing a comprehensive set of analyte specific codes for each unique non-culture dependent method for infectious agent detection. This code does not represent a new technology, and we believe it should also be cross walked to another single "infectious agent by immunofluorescent technique" code at 100% of the NLA. As this trend continues, it is also important to communicate to laboratory providers the importance of monitoring commercial kit products for FDA clearance status.

Codes 878XX

Infectious agent antigen detection by immunoassay with direct optical observation:

Streptococcus group B, Clostridium difficile toxin A, Influenza

Nature of the test method

Codes in the series for "infectious agent detection by immunoassay with direct optical observation" (87810-87899) were implemented in 1998 for use primarily with a specific "novel molecular thin film" based technology marketed by a single company. The technology, known as "optical immunoassay (OIA)" is based on the direct visually-apparent physical change that occurs in the thickness of a film due to binding reactions between antigens from clinical specimens and the antibodies used for antigen capture and for subsequent detection with the reactions occurring on a silicon chip. The tests are sold as kits for specific organism detection and include several common steps:

  1. Extraction of the infectious agent antigen from a clinical specimen. The actual extraction process varies depending on the specimen type.
  2. Mixing of the extracted sample with a conjugated antibody reagent.
  3. Transfer of the resulting mixture into the test cassette.
  4. Washing to remove excess reagents.
  5. Application of a substrate to enhance visual detection.

It appears that this series of codes has also gained acceptance for use with other enzyme immunoassay (EIA) based products that are interpreted by "direct optical observation," rather than using the multistep or single step EIA codes. For example, some companies recommend use with products based on an immunochromatographic approach. It is of note that some of these products for detection of Group A Streptococcus (87880) are waived under the Clinical Laboratory Improvement Amendments of 1988 (CLIA).

The new codes submitted for 2002 represent product line extensions for the "optical immunoassay" (OIA) method. In keeping with the current analyte and method specific coding approach for non-culture dependent infectious agent detection, each new analyte specific product will necessitate the establishment of a new code.


The applications of this technology are for the rapid qualitative detection of specific antigens in specific clinical specimens. The new codes proposed will add 3 additional etiologic agents to the existing specific analyte list. The new agents and some unique considerations of each include:

Streptococcus Group B: The performance characteristics of this product may not preclude the necessity for selective cultures to assure adequate sensitivity in the setting of the Centers for Disease Control and Prevention (CDC) recommended prenatal screening for this agent in cervicovaginal and rectal swabs. In this regard, use of the test is similar to that for the Group A Streptococcus antigen detection tests in which negative results should be confirmed by culture. Clearly, each individual institution will need to validate performance characteristics and establish medical staff or client approved reflex protocols for culture confirmation, if indicated.

  • Clostridium difficile Toxin A: The performance characteristics of this product may not preclude the necessity for additional testing using an assay that detects Toxin B since recent data indicate that strains of C. difficile that produce Toxin B only may be emerging as causes of antibiotic-associated diarrhea and colitis. Individual institutions will need to establish prevalence of these strains in their own settings to determine the best approach to diagnosis. If additional testing is done as a reflex protocol, medical staff or client approval will be necessary as described above.
  • Influena: This product is a multianalyte test that detects both major types of influenza virus, but does not distinguish between type A and B. Further, the performance characteristics of this product may not preclude the necessity for additional testing by another method (cell culture or antigen detection by immunofluorescent technique) in order to provide optimal sensitivity for detecting influenza infections in patients in a particular community. Thus, individual institutions will again need to evaluate performance in their own settings to determine the best approach and if additional testing is done as a reflex protocol, medical staff or client approval will be necessary.


In general terms, the total direct costs of performing these tests should be considered roughly equivalent, although volume of testing and the control of the OIA technology by a single vendor will clearly influence the actual reagent costs. At present, however, the NLA is uniformly set for all current codes including those modified for use in a CLIA-waived environment (modifier -QW).

Recommendations for reimbursement determination

These codes all represent a simple extension of the specific analyte listing for a specific immunoassay format previously deemed novel enough to justify a new method descriptor. They do not represent a new technology, and we believe the codes should be cross walked to the same uniform reimbursement level previously established at 100% of the $16.58 NLA.


Infectious agent genotype analysis by nucleic acid (DNA or RNA);

Hepatitis C virus (HCV)

Nature of the test method

Genotyping is a highly sophisticated technique to evaluate a microorganism for the presence of specific nucleic acid sequences. The sequences which are analyzed have been shown to correlate with specific organism characteristics that, in turn, influence clinical features of patients infected with that organism. The methods for performing genotyping vary but have in common a number of procedural steps:

  1. Amplification of one or more specific target nucleic acid sequences by a method such as the polymerase chain reaction.
  2. Purification of the amplified product in preparation for further molecular analysis.
  3. Molecular comparison of the amplified sequence to known sequences to determine the "relatedness" of the isolate in question to isolates with known clinical characteristics.

The methods vary primarily in the last step of this general process. At present, there are several commercially available products to perform this type of analysis. The two most common methods are described below:

  1. One method relies on bidirectional automated DNA sequencing. The complete procedure may be coded using separate CPT codes for each step in the procedure taken from the CPT series for molecular diagnostic techniques (83890-83912).

    The recommended code set includes:

    83891 x 3 (isolation or extraction of highly purified nucleic acid)

    83898 x 2 (amplification of patient nucleic acid, single primer pair, each pair)

    83904 (mutation identification by sequencing, each segment)

    83912 (interpretation and report)

    Note that the "home brew" approaches are generally modifications of this method, but may use slightly different code combinations to account specific steps in the each approach.

  2. The second method is termed a line probe assay. This is a reverse hybridization (also known as reverse "dot blot") test using specific probes immobilized in parallel lines on membrane strips to detect specific nucleic acid sequences. The recommended code set for this technique is quite different from the sequencing approach and includes:

87522 (Infectious agent detection by nucleic acid, DNA or RNA, Hepatitis C, quantification, a code which includes the initial amplified probe technique)

83896 x 21 (Molecular diagnostics; nucleic acid probe each).

It should be recognized that there are other approaches which may be coded in a substantially different manner. Further, there are currently no available FDA cleared products for genotyping, so users must be aware of current status for billing purposes. Users of products in an ASR format must also adhere to the regulations for use of ASR's as well as maintain awareness of other coverage issues.


Genotyping initially entered the mainstream of laboratory testing for determining whether patients infected with HIV-1 harbored molecular variants responsible for clinical resistance to antiretroviral agents. The technology was acknowledged by the implementation of a specific CPT code (87901) in 2001. Coincident with the use of genotyping to assess a patient's probable response to a specific "highly active anti-retroviral therapy" (HAART) regimen, the technology also gained popularity in managing patients with HCV infection. HCV can be classified into at least 6 major genotypes and more than 90 subtypes which show significant variance in geographic distribution. Further, it has been well established that differences exist in responses to antiviral therapy according to genotype. Genotype 1, which represents approximately 70% of cases in the United States, is associated with a poor treatment response and therefore worse outcome. Genotype 1 also shows a greater association with increased severity and a higher incidence of progression to liver cancer. Thus, genotyping plays a documented role in determining how to proceed with therapy as well as a prognostic role in assessing outcome.


Given the complexity of this test method, it should anticipated that associated costs are generally very significant. Each technique requires complex, expensive molecular diagnostic equipment, as well as costly molecular reagents and highly qualified laboratory clinical laboratory scientists. Our members indicate that direct costs per reportable result using a variety of methods are in the $250 to $300 range. Published reference laboratory prices, also performed by a variety of commercially available or "home-brew" techniques, are in the $350 to $400 price range.

Recommendation for reimbursement determination

Based on the definition of cross walking, ASM cannot support the concept of "equivalence" to CPT code 87901 for HIV genotyping. The reimbursement for 87901 was determined by cross walking to a composite set of codes that represented the use of a sequencing-based analysis for determination of genotype, which is in fact, the usual method for HIV-1 genotyping. However, as has already been mentioned, the composite code set for HCV genotyping shows marked variation based on specific approach.

We believe a gap fill approach is appropriate since there is not a single uniform composite code set which describes all methods for HCV genotyping. While it is probably appropriate to institute a single code since the end result provides data used in a clinically identical manner, the reimbursement amount is best based on the current gap fill method relying on regional charge data. This situation is not unlike that encountered with molecular amplification to detect infectious agents where there are a number of different amplification technologies, all of which fall under a single code for amplified probe for a specific infectious agent. CMS should, however, take the opportunity to collect data on exact methods used in this gap fill process, an undertaking which would be compatible with recommendation number twelve of the recent Institute of Medicine (IOM) Report on Medicare Laboratory Payment Policy, "to collect data needed to effectively manage the performance of the Medicare outpatient clinical laboratory payment system."


High sensitivity C-reactive protein

Nature of the test

C- reactive protein (CRP) is a non specific marker of an ongoing inflammatory process which can be specifically detected both qualitatively and quantitatively using an antibody-based immunoassay. Older methods employed latex agglutination with manual interpretation. As volumes increased, this method was often replaced by an automated latex particle- enhanced nephelometry method which is also used to detect other inflammatory markers such as haptoglobin. The "high sensitivity" CRP assays have been modified to reliably detect a lower limit of analyte which was previously considered in a "normal range." The methods currently used include latex-particle enhanced ephelometry and enzyme immunoassay.


The standard range CRP tests are used primarily to provide evidence for an active inflammatory process in the setting of a suspected infection or in the case of a suspected rheumatological disorder. In these cases, a physician seeks evidence that the CRP level is greater than the normal levels. In contrast, high sensitivity CRP tests are used to provide evidence for a low level inflammatory process to assess an individual's risk for heart disease. This application has arisen as a result of the rapidly emerging evidence for an infectious "trigger" for coronary artery disease. In these cases, physicians need to reliably measure CRP levels that are lower than those seen in acute infections.


The cost of performing high sensitivity CRP testing by latex particle-enhanced nephelometry is comparable to performing the standard assay. Likewise, measurement of CRP in a standard range has costs comparable to a high sensitivity range using an enzyme immunoassay format. Further, costs for other analytes by either method are also roughly comparable. In general terms, the high sensitivity CRP assay can be categorized as similar in costs based on methods to other protein assays with low measurement ranges (e.g. serum IgE, IgG subclasses, urine microalbumin, automated anti streptococcal antibody (ASO), and rheumatoid factor).

The critical reimbursement issue is that current NLA's set maximum payment for the specific CRP analyte (86140 at $7.15) at a substantially reduced amount compared to other similar analytes (haptoglobin, 83010 at $17.38 or beta 2 microglobulin, 82232 at $22.36). For the analytes described as similar by methodological approach, NLA's show significant disparity from ASO screen, 86063 at $7.98, to serum IgE, 82785 at $22.76. Further, for generic method codes payment is also higher, e.g. particle agglutination (86403 at $14.08); fluorescent noninfectious agent antibody; screen, each antibody (86255 at $16.66); immunoassay for analyte other than infectious agent antibody or infectious agent antigen, qualitative or semi-quantitative; multiple step method (83516 at $15.95); or immunoassay for analyte other than infectious agent antibody or infectious agent antigen, qualitative or semi-quantitative, single step method (83518 at $11.72). Compounding the issue with immunoassays for non-infectious analytes is the failure of the current CPT-4 system to consider both analyte and method in coding as is done for coding of microbiology non culture-dependent methods. If an analyte specific code has been assigned, one must accept the NLA for the analyte although methodology may, in fact, have significant impact on costs.

Recommendation for reimbursement determination

Although it would seem logical on the basis of analyte equivalence to cross walk high sensitivity CRP to the standard code, 86140, the underlying issues of previous failure to acknowledge cost considerations for specific methods leads us to conclude that this code should be a candidate for a gap fill process. We would suggest that CMS also collect information on methodology in this process to comply with recommendation number twelve in the IOM report on Medicare Laboratory Payment Policy to "collect data needed to effectively manage the performance of the Medicare outpatient clinical laboratory payment system."

Concluding comments

ASM appreciates the opportunity to offer comments regarding the assignment of payment rates for specific new CPT-4 codes for the 2002 Clinical Laboratory Fee Schedule. We commend CMS for moving toward a more open process as mandated by the Benefits Improvement and Protection Act (BIPA) and based on the IOM's report on Medicare Laboratory Payment Policy which recommends public consultation on reimbursement decisions. We are confident that such an open process will allow the successful coordination of science, business, and program integrity interests to ensure quality laboratory medicine for Medicare beneficiaries. ASM stands ready to work with you on this process.