P. Martinez et al., COMPARISON OF 2 ASSAYS FOR DETECTION OF HIV ANTIBODIES IN SALIVA, European journal of clinical microbiology & infectious diseases, 14(4), 1995, pp. 330-336
The presence of HIV antibodies was screened in 241 paired samples of s
erum and saliva from seronegative subjects with risk factors for human
immune deficiency virus (HIV) infection (n = 99), asymptomatic and sy
mptomatic HIV-seropositive patients (n = 104) and healthy blood donors
(n = 38) as negative controls, in order to assess the reliability of
two saliva tests for the detection of HIV antibodies. These results we
re confirmed by Western blot. The saliva samples were collected using
an oral device (Salivette) maintained in the lateral gingival fold unt
il the individual perceived that it was becoming less rigid due to hyd
ration with saliva. The two tests were a rapid one (Test Pack) and a c
onventional one (GACELISA). The results for antibody detection in sali
va show 99 % specificity and 99 % sensitivity for the rapid test versu
s 100 % sensitivity and 81 % specificity for the conventional test. Al
l results for the saliva samples which were positive in the rapid test
were confirmed by Western blot (WHO criteria), and there were no inde
terminate Western blot results among the samples which were false-posi
tive in the conventional enzyme immunoassay. No statistically signific
ant differences were observed between the absorbance values of HIV-inf
ected symptomatic and asymptomatic patients. The correlation for the r
esults of the HIV-antibody analysis in the paired sera was 98 %. This
method of saliva sampling in combination with a rapid and sensitive te
st for HIV-antibody detection may be applicable in studies conducted w
ith limited technical resources or insufficiently trained health perso
nnel or where blood sample collection is difficult.