Human T-lymphotropic virus type I (HTLV-I), the first human retrovirus to be identified, is associated with adult T-cell leukemia/lymphoma, HTLV-I-associated myelopathy/tropical spastic paraparesis (HAM/TSP), polymyositis, infective dermatitis, arthritis, and uveitis (1) (2-5) (6) (7) (1) (1)
Both the HTLVs are known to be transmitted by sexual contact (8-12) (8-10) (8,10) (13,14) tax antibody (13) (9) (15)
The National Heart, Lung, and Blood Institute's REDS includes HTLV-infected volunteer blood donors at five blood centers in Los Angeles, San Francisco, Baltimore/Washington, DC, Detroit, and Oklahoma City (16)
METHODS
Enrollment of Subjects
Details of the screening of volunteer blood donors for HTLV-I and enrollment of HTLV-seropositive donors in REDS have been described previously (16,17) (6)
Questionnaire of HTLV-Infected Donors and Their Sex Partners
At enrollment, each HTLV-infected donor was administered a questionnaire seeking information about behavioral risk factors, including drug use and sexual practices. Each donor was also asked about specific behaviors in the current, long-term (≥6 months) relationship, including duration of relationship and frequency of vaginal, oral, and anal intercourse. Total episodes of intercourse were calculated on the basis of the duration of the relationship and the reported frequency of intercourse. Subjects were also asked about the occurrence of sexually transmitted diseases (STDs) during their relationship. Sex partners were given the same questionnaire as the infected donors.
HTLV Serology
Serum specimens for donors and partners were tested for HTLV antibodies by enzyme-linked immunosorbent assay (ELISA), Western blot (WB), and radioimmunoprecipitation assay (RIPA) as previously described (17) gag p24 and env gp46 or gp61/68 by WB or RIPA were considered seropositive for HTLV-I/II (6) (18-23) env and whole virus antibody titers (Env1/5, MTA-1, RE3, Env2/20, K-55, and p21e-spiked whole virus) were expressed as reciprocals of the highest dilution that gave positive results after twofold dilutions, starting from 1/20. In all assays, the laboratory was blinded to the origin of the serum specimens.
PCR Confirmation and HTLV Subtyping
PCR was performed on lysates of peripheral blood mononuclear cells (PBMCs) from donors and partners (regardless of serologic status) as described (17) tax/rex and pol regions; HTLV-I was distinguished from HTLV-II by amplification in the pol region and selective hybridization. Subtyping of HTLV-I and HTLV-II infections was performed on PCR-amplified proviral DNA in the LTR region by restriction fragment length polymorphism (RFLP) as previously described (24,25)
PCR Analysis of Viral Load in Infected Male Partners
Quantitative PCR was performed on lysates of PBMCs at a concentration at 6 × 106 cells per mL, using a modification of the method described by Wood et al. (26) tax/rex ) primers were used. For HTLV-II, DC43 (TGGATACCCCGTCTACGTGT) and biotinylated CD44 (GAGCTGACAACGCGGTCCATCG), (tax/rex ) primers were used. Briefly, positive control lysates (MT-2 or MOT) were diluted with lysates of known negative donors to achieve a concentration of 100, 10, 1, and 0.1 positive control cells in 3 × 105 negative donor cells per 50 μL of lysate. Standard curve material consisted of the plasmid pMT-2 for HTLV-I and plasmid pH6B3.5 for HTLV-II in a dilution series representing 3 × 104 , 104 , 3 × 103 , 103 , 3 × 102 , 102 , 30, 10, and three copies in 3 × 105 normal/negative cells per 50 μL. Following amplification in a 9600 Gene Amp PCR system (Perkin-Elmer Cetus), the PCR product was added to an avidin-coated well, together with an HRP-probe (SK 45 for HTLV-I and DC 45 [ATGCCCTCCTGGCCACCTGTCCAGAGCACCAACTCACCTG] for HTLV-II). After washing, an OPD substrate (Sigma) was added, and the reaction was stopped by the addition of 1 N sulfuric acid. The optical density was read at 490 nm using a microplate reader (Coulter Corp.), and the copy number was calculated with “Soft-Max” software (Molecular Devices Corporation), using a loglogit fit of the standard curve material. Results were expressed as mean copy number per 300,000 total cells. For calculation, samples in which virus could not be detected in this assay were scored as 30 copies per 300,000 cells, the lower limit of detection in the assay. As in other assays, the laboratory was blinded to the sources of the specimens tested.
Analysis of Risk Factors for Male-to-Female Transmission
To investigate the likely direction of transmission in couples in whom both partners were infected, we assessed “major” (blood transfusion, born/sex partner from Caribbean, born/sex partner from Japan, IDU, sex partner of IDU) and “minor” (history of STDs, ≤10 lifetime sex partners, exchange of money or drugs for sex) behavioral risk factors according to the questionnaire data. We assessed the direction of transmission as “probable male-to-female” if 1) the female partner had no risk factor and the male partner had at least one major or at least two minor risks, or 2) the female partner had (at most) one minor risk, but the male had at least two risks, including at least one major risk. “Probable female-to-male” transmission was defined by reversing these requirements. For concordantly positive couples not satisfying the above criteria, the direction of transmission was designated “uncertain.”
To assess risk factors for male-to-female transmission (see RESULTS), we compared transmitter couples with non-transmitter couples by analyzing questionnaire data pertinent to sexual behavior in the relationship. When information (e.g., duration of relationship, frequency of sexual intercourse) was provided by both the male and female partners, we compared transmitter couples with non-transmitter couples by using both the male and the female responses. To assess laboratory markers for male-to-female transmission, we compared viral load and various anti-tax and anti-env antibodies in the male transmitters and male non-transmitters separately for HTLV-I and HTLV-II.
Statistical Analysis
In univariate analyses, dichotomized variables were compared using the chi square or Fisher's exact tests. Continuous variables were compared using the Wilcoxon 2-sample test, because these variables were generally not normally distributed (27) (28) (27)
RESULTS
HTLV Seropositivity and Direction of Transmission
A total of 546 HTLV-seropositive donors were enrolled in the study. Of these, 382 (70.0%) reported having a current sexual relationship of ≥6 months duration—113 men and 269 women. Of these, 40 men (35.4%) and 85 women (31.6%) brought their partners in to enroll in the study. One female donor brought a partner of the same sex to enroll in the study; this couple was eliminated from further analysis.
Of the 40 female partners of male donors, 15 (37.5%) were seropositive for HTLV; of the 84 male partners of female donors, 17 (20.2%) were seropositive. In all cases of concordant seropositivity, the HTLV type of the partner matched that of the donor, as determined by PCR. Additionally, subtyping of PCR products in the LTR region by RFLP showed a total of 5 HTLV-I and 4 HTLV-II subtypes in the study population; among 8 couples concordantly positive for HTLV-I and 16 couples condordantly positive for HTLV-II in which the virus subtypes in both partners could be determined, the subtypes in the male and female partners matched in all instances (data not shown). All partners who were seronegative were also PCR-negative, with the exception of one partner of an HTLV-I-seropositive man. This woman was HTLV-I ELISA-nonreactive and PCR-positive for HTLV-I. However, subsequent serologic testing of the initial specimen indicated reactivity in a p21e-enhanced HTLV-I ELISA and seropositivity by Western blot. Therefore, this woman was, in fact, HTLV-seropositive, and she is included in our analysis as an HTLV-I-infected partner.
Of the 32 concordantly positive couples, direction of transmission was assessed as “probable male-to-female” in 16 couples (6 HTLV-I-infected and 10 HTLV-II-infected) and “probable female-to-male” in 1 couple (HTLV-I-infected), and direction “uncertain” in the remaining 15 couples (5 HTLV-I-infected and 10 HTLV-II-infected). Because only one instance of “probable female-to-male” transmission was identified, we included the 15 concordantly positive couples in whom direction could not be assigned as instances of “possible male-to-female” transmission, and we limited further analysis to male-to-female transmission, including only the 31 “transmitter” couples (11 infected with HTLV-I and 20 with HTLV-II) and the 25 male-positive/female-negative “nontransmitter” couples (8 infected with HTLV-I and 17 with HTLV-II).
Risk Factors for Male-to-Female Transmission: Univariate Analysis
HTLV “transmitter” males were more likely to be older, to have been in their relationships longer, and to have had more total and unprotected episodes of vaginal sex with their partners than were “non-transmitter” males (Table 1) . Transmitter males also reported more episodes of sex during menses and sex causing bleeding; they appeared more likely to report a history of various STDs (syphilis, urethritis, and herpes) during their relationships, but none of these differences were statistically significant (Table 1) . The female partners of the transmitters (infected females) also reported having been in their relationships longer and having more episodes of vaginal sex than partners of non-transmitters, although only the former difference was statistically significant (215 months vs. 122 months, p = 0.005; other data not shown). Infected females also reported more episodes of sex during menses (204 vs. 137) and appeared more likely to report a history of herpes during their relationships (10% vs. 0%), but these differences were not statistically significant. Despite smaller sample sizes, most of the differences that were statistically significant in Table 1 were also significant when analyzed separately for HTLV-I and HTLV-II (4 of the 5 and 3 of the 5, respectively; data not shown).
Analysis of quantitative PCR results showed striking associations between transmitter status and viral load in both HTLV-I- and HTLV-II-infected males (Fig. 1) . Among the 16 HTLV-I-infected males in whom viral load was determined, those with the 5 highest and 4 lowest counts were transmitter and non-transmitter males, respectively. The geometric mean copy numbers/300,000 cells in transmitters and non-transmitters were 257,549 and 2,945, respectively. Similarly, among 35 HTLV-II-infected males, those with the 7 highest and the 13 lowest counts were transmitters and non-transmitters, respectively. The geometric mean copy numbers were 5,541 and 118, respectively. The associations between transmitter status and viral load were statistically significant for both viruses (p = 0.03 for HTLV-I and p < 0.001 for HTLV-II). Subtyping of the viruses revealed a total of five HTLV-I and four HTLV-II subtypes; HTLV-I and HTLV-II transmitter males did not appear more likely than non-transmitter males to be infected with any particular HTLV subtype (data not shown).
HTLV-I transmitter males appeared more likely than non-transmitter males to have anti-tax antibody (70% vs. 29%, p = 0.15), and to have higher antibody titers against p21e-spiked whole virus and immunodominant envelope epitopes represented by MTA-1, RE3, and Env1/5, although none of these differences achieved statistical significance (Table 2) . HTLV-II transmitter males demonstrated higher antibody titers against p21e-spiked whole virus and immunodominant envelope epitopes represented by K-55 and Env2/20; all of these differences were statistically significant (Table 2) .
Risk Factors for Male-to-Female Transmission: Logistic Regression Analysis
Logistic regression analysis of age and behavioral variables (using male responses because viral load and antibody titers were also measured in men) for both viruses combined indicated that, among the variables associated with male-to-female transmission in univariate analyses (duration of relationship, male age, and total episodes of vaginal sex), none could be shown to be a risk independent of the others. However, duration of relationship appeared to be the strongest predictor and was therefore entered initially into additional logistic regression models. For both HTLV-I and HTLV-II, viral load was more highly associated with male transmitter status than anti-tax or anti-env or whole virus antibodies; viral load was therefore entered initially into the models.
In a model containing log of duration of relationship, log of viral load, and HTLV type, both duration of relationship and viral load predicted risk for male-to-female transmission (Table 3) . The magnitude of the odds ratios suggests that a doubling of either duration of relationship or viral load approximately doubles the risk of male-to-female transmission (Table 3) . The high odds ratio (OR) for HTLV type (OR = 54.6) predicts that for a given viral load, an HTLV-II-infected man has a greater likelihood of being a transmitter than an HTLV-I-infected man. However, the 95% confidence interval (CI) around this odds ratio is wide (CI, 2.03-1,469, Table 3 ).
The duration of relationship and viral load parameters of the model were also estimated for HTLV-I and HTLV-II separately (i.e., the model included duration of relationship and viral load, but not HTLV type); similar trends were observed for duration of relationship and viral load, but the results were not as statistically significant because of smaller sample sizes (OR = 2.54, p = 0.106; and OR = 1.36, p = 0.084 for HTLV-I; OR = 3.56, p = 0.094; and OR = 3.49, p = 0.022 for HTLV-II). To evaluate the combined data further, additional models were estimated by replacing duration of relationship with total episodes of vaginal sex and male age, and viral load with various HTLV antibody results; none of these models improved on that using duration of relationship and viral load (data not shown).
Correlation Between Antibody Titers and Viral Load
Correlations were observed between various antibody titers and viral load for both HTLV-I and HTLV-II (see Table 2 ). For HTLV-I, statistically significant correlations were observed for p21e-spiked whole virus, MTA-1, RE3, and Env1/5 (see Table 2 ). For HTLV-II, statistically significant correlations were observed for two of the three antibody assays: K-55 and Env2/20 (see Table 2 ). A small number of HTLV-II-infected men had high antibody titers but low viral loads; this was particularly true for antibody to p21e-spiked whole virus (data not shown).
DISCUSSION
The findings from this study confirm and extend our knowledge regarding sexual transmissibility of the HTLVs. They support the hypothesis that the HTLVs are more efficiently transmitted from male to female than vice versa. Of 17 concordantly positive couples in whom directionality of transmission could be determined on the basis of risk factor assessment of both partners, 16 (94%) were probable cases of male-to-female transmission, whereas only 1 (6%) could be classified as “probable female-to-male” transmission. Similarly, the proportion of female partners of infected male donors who were themselves infected (37.5%) was nearly twice the proportion of male partners of infected female donors who were also infected (20.0%). In addition, the ratio of transmitter to non-transmitter couples was similar for HTLV-I and HTLV-II, suggesting similar efficiency of transmission of the two viruses.
Our study provides strong evidence that high viral load is an important determinant of male-to-female sexual transmission of both HTLV-I and HTLV-II. The relatively small overlap between viral loads in transmitter males versus non-transmitter males for both HTLV-I and HTLV-II was striking (see Fig. 1 ). Moreover, the statistically significant differences between viral loads in transmitters versus non-transmitters were noteworthy given the relatively small sample sizes in our study, particularly for HTLV-I. To our knowledge, these are the first data available concerning the direct relationship of viral load to transmission of the HTLVs, although such a relationship has been inferred in a study of mother-to-child transmission of HTLV-I (29) Fig. 1 ). However, the latter finding is unexplained and must be interpreted with caution; other investigators have not found levels of HTLV-I in asymptomatic persons as high as those found in this study (30,31)
Interestingly, few data are available concerning the relationship between viral load and transmission for the more intensely studied human immunodeficiency virus (HIV) and related lentiviruses. However, available data parallel our findings. Weiser et al. reported that high viral load correlates with mother-to-child transmission of HIV (32) (33) (34) (35) (36)
Associations between male transmitter status and HTLV antibody titers were significant only for HTLV-II-infected men, but the magnitude of the differences between antibody titers in HTLV-I transmitter versus non-transmitter males suggests that a larger sample size might have yielded more statistically significant results (see Table 2 ). In general, antibody titers correlated highly with viral load, as reported previously for HTLV-I (30,31,37) p = 0.198) may be related to the fact that the latter antigen is HTLV-I-derived. In logistic regression analysis, antibody titers were not as strongly associated with male-to-female transmission as viral load. These results suggest that antibody titers in the male partner are useful markers for male-to-female sexual transmission, put probably less so than viral load.
We also found that male-to-female transmission was associated with duration of relationship for both HTLV-I and HTLV-II, suggesting that repeated sexual contact increases the likelihood of viral transmission. Univariate analysis showed that transmission was also associated with total episodes of vaginal sex, but inclusion of this variable in logistic regression models did not improve on the association with duration of relationship; this finding is probably explained by the fact that total episodes of sex were calculated on the basis of the duration of relationship. Male-to-female transmission was also associated with male age, but we could not differentiate the effect of duration of relationship and age in our models.
An additional finding in our study was that serology by itself is a useful marker for infection in sex partners of HTLV-infected persons. We found only one instance in which a partner was seronegative on initial screening by ELISA, but PCR-positive for HTLV. However, subsequent testing of the same specimen indicated reactivity in a p21e-enhanced HTLV-I ELISA and seropositivity by Western blot. Thus, serologic testing of sex partners of HTLV-infected persons appears to be sufficient.
Several limitations in our study must be noted. Only one third of the HTLV-seropositive donors brought their sex partners in for enrollment in the study, and the enrolled couples may not be representative of all male-to-female transmitter and non-transmitter couples in the study population. The relatively low numbers of partner pairs, particularly for HTLV-I, made it difficult to examine simultaneously multiple behavioral risk factors for male-to-female sexual transmission. Therefore, we cannot conclude that variables that were not significant in our analyses, such as a history of STDs (9)
Of the 31 concordantly positive couples we included as instances of male-to-female transmission, risk factor information supported this designation in 16 cases; we cannot be certain that the male transmitted infection to the female in these or in the other 15 couples included. However, our inclusion of all these couples in the analysis is supported by several observations. First, the likelihood that transmission occurred between partners is supported by the subtyping data, which show concordance of subtypes in all couples. Second, as noted, male-to-female transmission appeared much more common than female-to-male transmission in this study population. Third, to investigate this issue further, we repeated the analyses in Table 3 restricting to “probable” and to “possible” male-to-female transmitters, respectively. The odds ratios and p values in these analyses were remarkably similar (OR = 2.37, p = 0.069; and OR = 2.33, p = 0.007 for duration of relationship and viral load for “probable” cases; and OR = 2.40, p = 0.054; and OR = 1.86, p = 0.011 for “possible” cases, respectively). When plotted visually, “probable” and “possible” male-to-female transmitters were also more similar to each other than to non-transmitters with regard to these variables (data not shown). All of these observations support the notion that most of these couples were cases of male-to-female transmission. Perhaps most importantly, any misclassification of these cases would have made it more difficult to demonstrate the associations we found with these small sample sizes.
Finally, serum and PBMCs were collected after viral transmission occurred in the transmitter couples, and the antibody titers and viral loads we measured may not be representative of those at the time of viral transmission. The natural history of these variables in HTLV-infected persons and, therefore, the effect of this delay in our analyses, is unknown. Further, the conclusion that male-to-female transmission is associated with duration of relationship assumes that the male partner was infected when these relationships began; this may not be true for many of the couples in our study.
Our data provide new information and a biologically plausible framework in which to assess risk for male-to-female sexual transmission of the HTLVs. However, recommendations concerning counseling of HTLV-infected persons still apply (38) (38)
Acknowledgment: The authors thank Harold Jaffe for review of the manuscript, Kevin Watanabe for statistical support, John O'Connor for editorial assistance, Jean Smith for preparation of the figures, and Ruby Booth, Marcia Smalls, and Martha Gaillard for secretarial assistance. This work was supported by NHLBI Contracts NO1-HB-97077 (superceded by NO1-HB-47114), -97078, -97079, -97080, -97081, and -97082.
FIG. 1:
. Viral load (mean copy number/300,000 cells) in HTLV-I- and HTLV-II-infected males. Data available for 16 HTLV-I-infected males (9 transmitters, 7 non-transmitters) and 35 HTLV-II-infected males (19 transmitters, 16 non-transmitters). The geometric mean copy numbers/300,000 cells (solid lines) were 257,549 and 2,945 in HTLV-I transmitters and non-transmitters, respectively (p = 0.03); and 5,541 and 118 in HTLV-II transmitters and non-transmitters, respectively (p < 0.001, Wilcoxon 2-sample test). T = transmitters; NT = non-transmitters.
APPENDIX
The Retrovirus Epidemiology Donor Study (REDS) is currently the responsibility of the following persons: A. E. Williams (Holland Laboratory) and C. C. Nass, American Red Cross Blood Services Greater Chesapeake and Potomac Region; H. E. Ownby and A. W. Shafer, American Red Cross Blood Services Southeastern Michigan Region; S. H. Kleinman (UCLA Medical Center) and S. Hutching, American Red Cross Blood Services Southern California Region; E. L. Murphy (UCSF Medical Center) and M. P. Busch, Irwin Memorial Blood Centers; R. O. Gilcher and J. W. Smith, Oklahoma Blood Institute; G. B. Schreiber and R. A. Thomson, Westat, Inc.,; G. J. Nemo, National Heart, Lung, and Blood Institute, National Institutes of Health; and T. F. Zuck (Hoxworth Blood Center), Steering Committee Chairman.
REFERENCES
1. Kaplan JE, Khabbaz RF. The epidemiology of human T-lymphotropic virus types I and II.
Rev Med Virol 1993;3:137-48.
2. Hjelle B, Appenzeller O, Mills R, et al. Chronic neurodegenerative disease associated with HTLV-II infection.
Lancet 1992;339:645-6.
3. Jacobson S, Lehky T, Nishimura M, Robinson S, McFarlin DE, Dhib-Jalbut S. Isolation of HTLV-II from a patient with chronic, progressive neurological disease clinically indistinguishable from HTLV-I-associated myelopathy/tropical spastic paraparesis.
Ann Neurol 1993;33:392-6.
4. Harrington WJ, Sheremata W, Hjelle B, et al. Spastic ataxia associated with human T-cell lymphotropic virus type II infection.
Ann Neurol 1993;33:411-4.
5. Murphy EL, Engstrom JW, Miller K, et al. HTLV-II associated myelopathy in a 43-year-old woman.
Lancet 1993;341:757-8.
6. Centers for Disease Control. Licensure of screening tests for antibody to human T-lymphotropic virus type I.
MMWR 1988;37:736-40,745-7.
7. Centers for Disease Control. Human T-lymphotropic virus type I screening in volunteer blood donors—United States, 1989.
MMWR 1990;39:915,921-4.
8. Kajiyama W, Kashiwagi S, Ikematsu H, Hayashi J, Nomura H, Okochi K. Intrafamilial transmission of adult T cell leukemia virus.
J Infect Dis 1986;154:851-7.
9. Murphy EL, Figueroa JP, Gibbs WN, et al. Sexual transmission of human T-lymphotropic virus type I (HTLV-I).
Ann Intern Med 1989;111:555-60.
10. Stuver SO, Tachibana N, Okayama A, et al. Heterosexual transmission of human T-cell leukemia/lymphoma virus type I among married couples in southwestern Japan: an initial report from the Miyazaki cohort study.
J Infect Dis 1993;167:57-65.
11. Vitek CR, Gracia FI, Giusti R, et al. Evidence for sexual and mother-to-child transmission of human T-lymphotropic virus type II among Guaymi Indians, Panama.
J Infect Dis 1995;171:1022-6.
12. Hjelle B, Cyrus S, Swenson SG. Evidence for sexual transmission of human T-lymphotropic virus type II.
Ann Intern Med 1992;116:90-1.
13. Chen Y-MA, Okayama A, Lee T-H, Tachibana N, Mueller N, Essex M. Sexual transmission of human T-cell leukemia virus type I associated with the presence of anti-
tax antibody.
Proc Natl Acad Sci USA 1991;88:1182-6.
14. Ho GYF, Nomura AMY, Nelson K, Lee H, Polk BF, Blattner WA. Declining seroprevalence and transmission of HTLV-I in Japanese families who immigrated to Hawaii.
Am J Epidemiol 1991;134:981-7.
15. Schwebke J, Calsyn D, Shriver K, et al. Prevalence and epidemiologic correlates of human T-cell lymphotropic virus infection among intravenous drug users.
J Infect Dis 1994;169:962-7.
16. Kleinman SH, Kaplan JE, Khabbaz RF, et al. Evaluation of a p21e-spiked Western blot (immunoblot) in confirming human T-cell lymphotropic virus type I or II infection in volunteer blood donors.
J Clin Microbiol 1994;32:603-7.
17. Busch MP, Laycock M, Kleinman SH, et al. Accuracy of supplemental serologic testing for human T-lymphotropic virus type I and II in United States blood donors.
Blood 1994;83:1143-8.
18. Lal RB, Rudolph DL, Griffis KP, et al. Characterization of immunodominant epitopes of
gag and
pol gene encoded proteins of human T-lymphotropic virus type-I.
J Virol 1991;65:1870-6.
19. Rudolph DL, Lal RB. Discrimination of human T-lymphotropic virus type-I and type-II infections by synthetic peptides comprising structural epitopes from the envelope glycoproteins.
Clin Chem 1993;39:288-92.
20. Rudolph DL, Colligan JE, Lal RB. Detection of antibodies to
trans -activator protein (p40
tax ) of human T-cell lymphotropic virus type I by a synthetic peptide-based assay.
Clin Diag Lab Immunol 1994;1:176-81.
21. Buckner C, Roberts CR, Foung SKH, et al. Immune responsiveness to the immunodominant envelope epitopes of HTLV-I and HTLV-II in diverse geographic populations.
J Infect Dis 1992;166:1160-3.
22. Hino S, Katamine S, Miyamoto T, et al. Association between maternal antibodies to the external envelope glycoprotein and vertical transmission of HTLV-I: maternal anti-
env antibodies correlate with protection in non-breast-fed children.
J Clin Invest 1995;95:2920-5.
23. Hartley TM, Malone GE, Khabbaz RF, Lal RB, Kaplan JE. Evaluation of recombinant human T-cell lymphotropic virus type I (HTLV-I) p21E antibody detection enzyme immunoassay as a supplementary test in HTLV-I/II antibody testing algorithms.
J Clin Microbiol 1991;29:1125-7.
24. Switzer WM, Pieniazek D, Swanson P, et al. Phylogenetic relationship and geographic distribution of multiple human T-cell lymphotropic virus type II subtypes.
J Virol 1995;69:621-32.
25. Vidal AU, Gessain A, Mitsuyaki Y, et al. Phylogenetic classification of human T cell leukemia/lymphoma virus type 1 genotypes in five major molecular and geographical subtypes.
J Gen Virol 1994;75:3655-66.
26. Wood R, Dong H, Katzenstein DA, Merigan TC. Quantification and comparison of HIV-1 proviral load in peripheral blood mononuclear cells and isolated CD4+ cells.
J Acquir Immune Defic Syndr 1993;6:237-40.
27. Snedecor GW, Cochran WG, eds.
Statistical methods, 7th ed. Ames, IA: Iowa State University Press, 1980.
28. Homer DW, Lameshow S, eds.
Applied logistic regression . New York: John Wiley & Sons, 1989.
29. Yoshinaga M, Yashiki S, Oki T, Fujiyoshi T, Nagata Y, Sonoda S. A maternal risk factor for mother-to-child HTLV-I transmission: viral antigen-producing capacities in culture of peripheral blood and breast milk cells.
Jpn J Cancer Res 1995;86:649-54.
30. Matsumura M, Kushida S, Ami Y, et al. Quantitation of HTLV-I provirus among seropositive blood donors: relation with antibody profile using synthetic peptides.
Int J Cancer 1993;55:220-2.
31. Shinzato O, Kamihara S, Ikeda S, et al. Relationship between the anti-HTLV-I antibody level, the number of abnormal lymphocytes, and the viral-genome dose in HTLV-I-infected individuals.
Int J Cancer 1993;54:208-12.
32. Weiser B, Nachman S, Tropper P, et al. Quantitation of human immunodeficiency virus type 1 during pregnancy: relationship of viral titer to mother-to-child transmission and stability of viral load.
Proc Natl Acad Sci USA 1994;91:8037-41.
33. Busch MP (Transfusion Safety Study). Clinical and virological factors influencing transmission of HIV-1 by transfusion [Abstract].
Transfusion 1992;32(8S):56S.
34. Clerici M, Clark EA, Polacino P, et al. T-cell proliferation to subinfectious SIV correlates with lack of infection after challenge of macaques.
AIDS 1994;8:1391-5.
35. Cardo D, Srivasztava P, Ciesielski C, et al. Case-control study of HIV seroconversion in health care workers (HCWs) after percutaneous exposure (PE) to HIV-infected blood. Abstract No. 1236. Abstracts of the 34th Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington, DC: American Society for Microbiology, 1994.
36. O'Brien TR, Busch MP, Donegan E, et al. Heterosexual transmission of human immunodeficiency virus type 1 from transfusion recipients to their sex partners.
J Acquir Immune Defic Syndr 1994;7:705-10.
37. Ishihara S, Okayama A, Stuver S, et al. Association of HTLV-1 antibody profile of asymptomatic carriers with proviral DNA levels of peripheral blood mononuclear cells.
J Acquir Immune Defic Syndr 1994;7:199-203.
38. Centers for Disease Control and U.S.P.H.S. Working Group. Guidelines for counseling persons infected with human T-lymphotropic virus type I (HTLV-I) and type II (HTLV-II).
Ann Intern Med 1993;118:448-54.
