Discussion

Our study was designed to measure HIVl RNA in routinely processed clinical samples stored under conventional conditions.The combination of RNAzol plasma extraction and RT-PCR yields a reliable assay for HIV1 RNA，with results obtainable in less than two days.The assay is responsive to 2-fold changes in RNA copy number when this number is maintained within the linear range of the amplification protocol.The effect of this assay on recovery and stability of HIV1 RNA is the subject of a separate publication（Zhu et al.，1995）.Here we have investigated the clinical correlates of this technique in patients with HIV1 infection.

Using cumbersome culture methods，investigators have found that plasma viraemia is detectable in all clinical stages of HIV infection（Ho et al.，1989；Coombs et al.，1989）.Holodniy et al.（1991a）utilized a PCR method to quantitate viral load in serum.The authors reported that viral RNA copy numbers from HIV1-seropositive patients ranged from 500 to 250，000/ml.In our study，measurements of HIV1 RNA copies in plasma from patients at different stages of infection ranged from 800 to 500，000/ml（see Figure 1）.These results are similar to the findings in serum.In an effort to decrease the variability of RNA PCR，quantitative HIV1 RNA assays have recently been developed using the competitive PCR technique（Scadden et al.，1992；Menzo et al.，1992；Piatak et al.，1993）.Our results suggest，however，that the additional steps used in the competitive assay may not be necessary when a reliable RNA extraction method and a rigorous amplification protocol are employed.Further comparison of these techniques should be informative.

We examined the relationship between HIV1 RNA levels and clinical status，CD4+Tcell counts and AZT therapy in our patients.We found that plasma HIV1 RNA levels correlated roughly with the clinical stage of HIV disease，with levels above 25，000/ml indicating advanced disease［see Figure 2（a）］.To a lesser extent，HIV1 RNA levels also correlated with CD4+T-cell counts.Although these lymphocyte counts are considered to be useful surrogate markers in HIV disease，CD4+T-cell numbers may fluctuate for a variety of reasons during the course of HIV infection（Amadori et al.，1992）.Nevertheless，in serial samples from individual patients，the HIV1 RNA level paralleled the CD4+T-cell count in each case［see Figure 2（b）］.Thus quantitation of HIV1 RNA appeared to provide a consistent indicator of disease activity in individual cases.

We also found that patients treated with AZT tended to have higher levels of HIV1 RNA than patients on no therapy（see Table 1）.This difference probably reflects the more advanced clinical stage of patients on antiretroviral therapy.Previ ous culture studies have suggested that AZT may reduce the viral load in HIV-infected patients（Ho et al.，1989；Coombs et al.，1989）.In our serial and cross-sectional analyses，patients with more advanced HIV disease did not appear to respond significantly to AZT，as judged by direct measurement of viral load［see Figure 2（b）and Table 1］.However，our study was not designed to measure the in vivo effect of antiviral therapy，and no conclusions can be drawn from the RNA measurements in AZT-treated patients.(https://www.daowen.com)

It is of interest that every patient whose plasma had HIV1 RNA copies above 25，000/ml had ARC or AIDS.However，patients whose plasma HIV1 RNA copies were below this level were not necessarily at an early stage of HIV infection.One AIDS patient（No.23）who received topical treatment with the immune modulator DNCB had a relatively low level of HIV1 RNA（5，000 copies/ml）.A follow-up study has confirmed this initial observation of a decreased viral load in DNCB-treated patients（Stricker et al.，1993）.Two other AIDS patients（Nos.16 and 28）also had relatively low HIV1 RNA levels.One of these patients had recently started ddI therapy，which may have resulted in a decreased viral load（Holodniy et al.，1991b）.

Table 1　Relationship between plasma HIV1 RNA levels and AZT treatment

In conclusion，direct HIV1 RNA quantitation using RNAzol extraction and RT-PCR provides an accurate and reliable marker for HIV disease activity.This simplified assay should prove useful for monitoring the efficacy of various treatment modalities in HIV infection.