Abstract
Dear Sir,
Davidson 1 raises some interesting and valid points for discussion in response to our review article. 2 Characterization and typing of the cryoprecipitate should always be performed as the type of the cryoglobulin is strongly indicative of the most likely aetiology. In a type 1 cryoglobulin, one would look hard for a lymphoproliferative disease whereas a type 2 or type 3 cryoglobulin would be investigated differently. A cryoglobulin in the context of Sjogren's is very different from the one in hepatitis infection.
Is known hepatitis C virus (HCV) status a barrier to proper characterization? A laboratory that follows good practice in sample handling should be able to deal with category 3 samples and it should not be a deterrent to proper processing, characterization and typing of a cryoprecipitate. A significant number of unknown infectious samples are handled by any large laboratory every day, hence universal sample handling precautions. Knowledge of the potential risk status of an individual serum allows risk assessment, and in our laboratory we do not shy away from characterization on the basis of known HCV infection, but only where clinically relevant.
Davidson 1 states that the vast majority of cryoglobulin requests in their laboratory are from patients who are known to have HCV infection with a ‘simple question’ – ‘does this patient have a cryoglobulin which can explain their symptoms?’ The assumption is that many patients with HCV infection require this test on clinical grounds. Previous studies by one of us would suggest that >90% of infected patients do not, and those who have clinical symptoms have other clues to aetiopathogenesis, such as larger cryoglobulins and monoclonal components – making typing and some form of locally validated quantitation quite important, at least in the UK population studied. 3 Mixed cryoglobulins are reported in 20–56% of HCV-infected individuals, 4,5 but only 5–27% of the HCV cryoglobulins cause cryoglobulinaemic syndromes and there are usually other clues pointing to this pathology. 6 Complement levels vary in cryoglobulinaemia and are insufficient when used alone. As we clearly recommend in our review, cryoglobulin testing should be reserved for patients with appropriate clinical symptoms and should not be offered if there are no symptoms attributable to cryoglobulinaemia.
Good laboratory practice includes liaison with users to determine whether the testing strategy is appropriate; educating laboratory users where it is not; appropriate requesting and gating of tests; and once a test is done, proper interpretation.
In response to the valid questions raised by Davidson, 1 we would ask, what are the criteria used by the clinicians to request typing and quantitation? Are they selecting for high probability of cryoglobulinaemic pathology which makes the pretest probability sufficiently high to dispense with further characterization? We would suggest that most laboratories, testing predominantly for the presence of cryoglobulins in autoimmunity and lymphoproliferation, will have different requirements for additional characterization. We would posit that the different practice in Davidson's laboratory reflects the reality of pretest selection and does not invalidate the recommendations. Each laboratory must adapt their practice to the population screened.
We agree with Davidson's comments that cryoglobulins precipitating outside three days are unlikely to be clinically significant. However, while this appears common-sense, finding evidence to backup this view is difficult, and hearing anecdotal evidence of someone who apparently broached this rule is not uncommon in laboratories testing large numbers. We are not contradicting this view and hence our recommendation is at least for three days. We recommend incubation at 4°C for at least three days and the reference for seven days has been quoted in our review. 7
We agree that laboratory practice in the real world is constrained by resource limitations and good laboratory practice entails optimizing resource allocation. Best practice guidance is an attempt to encapsulate the current state of knowledge regarding optimum assessment of cryoglobulins, and all accreditation systems would require the individual laboratory to validate the approach to testing and reporting in their laboratory and their screened population. It has been shown in several studies that the aetiological basis of different types of cryoglobulin varies and, further, that the investigation and management of these conditions are entirely different. We conclude that even in cryoglobulinaemia associated with HCV infection, assessment of quantity, type and clinical relevance is important.
In our hospital, we do not offer cryoglobulin measurement in people who have chronic HCV or hepatitis B virus infection without symptoms potentially attributable to cryoglobulins. We stand by the view that a laboratory result stating that a ‘significant’ cryoglobulin is present, without any further interpretation, is essentially useless to the clinician.
DECLARATIONS