Wilson's theories about seroconversion are as similarly weak, as are most of his other theories. Here, Wilson tries to convince the reader that "poppers" increase the chance of seroconverting. However, as Dr. Steve Harris reminds us, among many other credible studies that also dispute Wilson, are the results of a huge San Francisco study, began in the mid-1980's, which clearly demonstrated that "poppers" appear to have no effect on seroconverting: "Readers will remember that once men were infected with HIV, subsequent use of poppers, and subsequent numbers of partners, made no impact on future risk of developing AIDS. That is one important way epidemiologists know they have the cause of a problem. If many things correlate with risk of getting a diseases, but all things stop correlating after one of the variables changes, then that is likely to be the causal variable of interest. We now have studies showing such a relationship between HIV and AIDS, and between KSHV and KS."
Steve Harris, M.D.
Ruiz, et al (1998) Risk factors for human immunodeficiency virus infection and unprotected anal intercourse among young men who have sex (YMSM) with men. Sexually transmitted diseases, 25: 100.
Ruiz relied on self reporting, the reliance of which from YMSM is unknown. He also did not distinguish between frequency of use, thus the analysis treated those having used a drug once the same as those who used that drug daily. He also grouped nitrite use with another stimulant, crack, which further confounds his results.
Although Ruiz found a slight positive association between recent use and recent UAI, the association between lifetime use and HIV infection is much larger. The first result would indicate no relationship between UAI and nitrite use and the second result is unexplained.
Chesney, et al (1998) Histories of substance use and risk behavior: Precursors to HIV seroconversion of homosexual men. Amer J Public Health, 88:113.
This is a report generated from data from surveying the San Francisco Men's Health Study cohort. Although they do not discuss the implications of this, history of consistent use of amyl nitrite or amphetamines strongly affected seroconversion, while current use of these drugs did not. The data does not support a role for nitrite use in seroconversion.
Chesney did not measure whether substance use occurred at the time of sexual activity that may have caused seroconversion, this is an important factor.
McFarland, et al. (1997) Estimation of human immunodeficiency virus (HIV) seroincidence among repeat anonymous testers in San Francisco. American Journal of Epidemiology, 146:662.
In this study, the investigators claimed that nitrite use was significantly associated with seroconversion. However, only 42 out of 789 men who reported nitrite inhalation during sex in the last year were HIV-positive. Since the incidence of seropositivity in nitrite users is so small, one cannot establish a statistical significance using this data.
Ostrow, et al. (1995) A case-control study of human immunodeficiency virus
type I seroconversion and risk-related behaviors in the Chicago MACS/COS cohort. American Journal of Epidemiology, 142(8):875.
This study found that seroconversion correlated with marijuana, cocaine, and nitrite, but not alcohol use. This contradicts other studies finding that a relationship of alcohol use with HIV infection. Furthermore, it was not reported if subjects used other drugs in conjunction with nitrates, which they likely did and that would confound the issue such that nitrite use can be directly correlated with seroconversion.
Seage, et al. (1992) The relation between nitrite inhalants, unprotected receptive anal intercourse, and the risk of human immunodeficiency virus infection. American Journal of Epidemiology, 135:1.
This article attempted to determine whether nitrite use is an independent risk factor for HIV infection and if it interacts with unprotected receptive anal intercourse to further increase that risk. They collected self-reports from homosexual male couples and performed a number of statistical tests to obtain odds ratios for associations between different behaviors and HIV infection.
The ratio of unprotected anal sex in participants who never used nitrites during sex (9.0) was higher than the number for those who sometimes used nitrites during sex (7.1), which refutes the theory that nitrite use leads to unsafe sex. Although the number for those who always used poppers during sex was higher (31.8), it is likely that a group that always uses any drug during sex may be a skewed population.
The odds ratio was similar when comparing the association of HIV seropositivity with use of several different recreational drugs (a range of 1.7 for cocaine use to 2.9 for nitrite use) or HIV seropositivity with nitrite use and unprotected anal sex. The similarity of these numbers implies that any type of drug use can be correlated to HIV seropositivity. The overall odds ratio for HIV infection in all study participants who used nitrites was only 1.6, which is very low.
An important control that was not included is the incidence of systemic bleeding during unprotected anal sex was not available. This could be the most important risk factor for HIV infection.
In their conclusion, they state that "It did appear that the relation between nitrite use and HIV infection was confounded, since the odds ratio decreased from 2.9 to 1.7 after we controlled for confounding variables as well the presence of a study partner. We suspect that the remaining increased risk associated with nitrite use results from residual confounding. We reanalyzed the data with nitrite use recoded into eight exposure categories using Rosner's model and found that nitrite use was no longer significantly associated with HIV infection." Thus, controlling for confounding factors eliminates the associations between nitrite use and unprotected sex.
Finally, it is stated in the article that this group has never found an association between nitrite use and Kaposi's sarcoma. These results refute the proposed hypothesis that nitrites have an epidemiological role in ks.
Messiah (1993) Factors correlated with homosexually acquired human immunodeficiency virus infection in the era of "safer sex". Sexually Transmitted Diseases, 20:51.
This study also has a disparate sample size (n=201 for seronegative participants and n=45 for seropositive participants). Secondly, although inhalation of nitrites was significantly related to seropositivity, upon multivariate analysis, there was no significant difference between the two parameters. Once again, after controlling for other risk factors, there is no correlation between nitrite use and seroconversion.
Penkower (1991) Behavioral, health and psychosocial factors and risk for HIV infection among active homosexual men: the multicenter AIDS cohort study. American Journal of Public Health, 81:194.
This report contains only one table and has the same flaw as the one that was previously discussed, which is that the number of seronegative (463) versus seropositive (181) participants is extremely disparate, making accurate conclusions difficult. Regardless of this limitation, the data presented indicates that there is a similar increase in seroconversion for alcohol, cigarette, and recreational drug use, with heavy alcohol consumption being the most strongly associated with subsequent seroconversion. Although the researchers claim that nitrite use had the highest risk factor for seropositivity of all other drugs tested, they do not list the drugs that were studied or show the data for these drugs. They also state that drug users were more likely to engage in anonymous sex and had more partners, which are definitely confounding factors.
Burcham et al. (1989) Incidence and risk factors for human immunodeficiency virus seroconversion in a cohort of Sydney homosexual men. The Medical Journal of Australia, 150:634.
In this study, the researchers state that the relative risk of seroconversion was significantly higher among subjects who abused nitrite inhalant during the seroconversion period and that there was a significant relationship between nitrite use and anal receptive intercourse. From these two correlation's, a causal relationship cannot be demonstrated between nitrite use and seroconversion. Furthermore, the author claimed that the drug use may have been a correlate of high-risk behavior and this is the only conclusion that can be accurately formed.
van Griesven et al.(1987) Risk factors and prevalence of HIV antibodies in homosexual men in the Netherlands, American Journal of Epidemiology, 125:1048.
Seropositive respondents took several more drugs than seronegatives, thus confounding results. Additionally, these drugs may lead to a more intensive contact and in that way to a higher probability of transmission and since there was more than one used, it is impossible to tell which one, if any, is responsible for seroconversion.
(1992) The relationship between nitrite inhalants, unprotected anal intercourse, and the risk of Human Immunodeficiency Virus Infection, American Journal of Epidemiology, 135:1.
The role of nitrite use was evaluated between 1984 and 1988 in a study of sexual transmission of HIV among homosexual male couples in Boston. Even though the OR for HIV and unprotected receptive anal intercourse was higher than the OR for HIV and nitrite use, they administered a supplemental questionaire to determine if nitrite use might be a marker for unprotected receptive anal intercourse. The OR for those who always used nitrites during unprotected receptive anal intercourse was 31.8, compared with men who sometimes used nitrites (OR=7.1) or never (OR =9). Their conclusion was “the results of this study suggest that use of nitrate inhalants interacts with unprotected receptive anal intercourse to increase the risk of HIV infection. This is faulty logic, primarily because the OR for those who do not use nitrites is higher that those who sometimes use them.
Also, the confidence intervals are high, ranging from 1-76.7. The width of the confidence interval gives an idea about the uncertainty of the unknown parameter. A very wide interval may indicate that more data should be collected before anything very definite can be said about the parameters.
