Skip Navigation

Primary Human Immunodeficiency Virus Type 1 Infection: Clinical Manifestations among Women in Mombasa, Kenya

  1. Ludo Lavreys1,2,
  2. Mary Lou Thompson1,
  3. Harold L. Martin Jr.1,a,
  4. Kishorchandra Mandaliya3,
  5. Jeckoniah O. Ndinya-Achola2,
  6. Job J. Bwayo2, and
  7. Joan Kreiss1
  1. 1Departments of Epidemiology, Medicine, and Biostatistics, University of Washington, Seattle, Washington
  2. 2Department of Medical Microbiology, University of Nairobi, Mombasa, Kenya
  3. 3Coast Provincial General Hospital, Mombasa, Kenya
  1. Reprints or correspondence: Dr. Joan Kreiss, University of Washington, Box 359909, 325 Ninth Ave., Seattle, WA 98104-2499.

  • ↵a Current affiliation: Park Nicollet Clinic, Minneapolis, Minnesota.

 
Next Section

Abstract

The occurrence of clinical manifestations associated with primary human immunodeficiency virus type 1 (HIV-1) infection was evaluated in a prospective cohort study of female sex workers in Mombasa, Kenya. Among 103 women who seroconverted to HIV-1, fever, vomiting, diarrhea, headache, arthralgia, myalgia, skin rash, swollen lymph nodes, extrainguinal lymphadenopathy, inguinal lymphadenopathy, and vaginal candidiasis were noted significantly more frequently at visits in which seroconversion first became evident. Eighty-one percent of seroconverting women had ≥1 of these 11 symptoms or signs. Among 44% of the women, the acute illness was severe enough to prevent them from working. Having ≥2 of 6 selected symptoms and signs yielded a sensitivity of 51%, specificity of 83%, positive likelihood ratio of 3.2, and negative likelihood ratio of 0.5 for acute HIV-1 infection. The recognition of primary HIV-1—infection illness in high-risk populations and subsequent risk-reduction counseling could potentially reduce secondary HIV-1 transmission during this highly infectious period.

Acute HIV-1 infection often results in a symptomatic mono-nucleosis-like illness. The most frequently reported symptoms and signs include fever, fatigue, rash, headache, lymphadenopathy, pharyngitis, myalgia, arthralgia, and gastrointestinal problems [1]. The illness is self-limited, but it may last up to 4 weeks and sometimes is of sufficient severity to warrant hospitalization. Symptomatic illness corresponds with an early phase of high viral load, and resolution of symptoms occurs as the host immune response evolves and HIV-1—specific anti-bodies become detectable.

The frequency of symptomatic primary HIV-1 infection is unknown [2]. Most published studies have been cross-sectional and have described clinical signs and symptoms among patients who came to medical attention. It has been estimated that 50%–90% of newly infected individuals are symptomatic, but this may be an overestimate, given the limitations of the study designs [1, 3]. In addition, the signs and symptoms that characterize primary HIV-1 infection are nonspecific, and few published studies have included a seronegative control group for comparison [4]. Comparing the frequency of signs and symptoms in both case and control patients would help determine their utility as screening tools for newly acquired HIV-1 infection.

Data on acute HIV-1 illness from developing countries are scarce. In a nested case-control study, Bollinger et al. studied patients with sexually transmitted diseases (STDs) in Pune, India, and found that a positive p24 antigen test, suggesting recent infection with HIV-1, was significantly associated with a history of fever, joint pain, and night sweats, after adjustment for presence of STDs [5].

In Mombasa, Kenya, we initiated a prospective study of a cohort of HIV-1—seronegative female sex workers in 1993 as part of the Preparation for AIDS Vaccine Evaluation (PAVE)/HIV Network for Prevention Trials (HIVNET) initiative, with the aims of determining HIV-1 seroincidence and correlates of seroconversion [6]. Over a 5-year period, clinical, behavioral, and laboratory data were prospectively collected, including data on symptoms and signs associated with acute HIV-1 illness, allowing comparison of their occurrence in seronegative and newly HIV-1 infected patients.

Previous SectionNext Section

Patients and Methods

Study patients and procedures. Between March 1993 and March 1998, sex workers who attended Ganjoni Municipal Clinic in Mombasa were invited to enroll in a prospective cohort study, as described elsewhere [6]. Women underwent serological testing for HIV-1 after providing informed consent. HIV-1—seronegative women were enrolled in the cohort study.

At the enrollment visit, data were collected regarding demographic variables, sexual behavior, and medical history. Patients underwent a general physical examination, including pelvic examination, and STD screening. Study patients were then asked to return every month for follow-up. At each visit, a questionnaire on interim sexual behavior and medical history was administered, covering the period since the last clinic visit. Specifically, women were asked about possible symptoms of primary HIV-1 infection. A physical examination was performed, and clinical signs associated with primary HIV-1 infection were noted. STD screening was conducted, and a blood sample for HIV-1 serological testing was obtained. All women received risk-reduction counseling and a free supply of condoms. Women with an STD received appropriate treatment.

Laboratory procedures. HIV-1 serological testing was performed with an ELISA (Detect-HIV; Biochem ImmunoSystem, Montreal). A second confirmatory ELISA (Recombigen; Cambridge Biotech, Worcester, MA) was performed if a sample tested positive on the screening ELISA. All HIV-1 seroconversions during follow-up were confirmed by HIV-1 Western blotting (Cambridge Biotech). Screening for Candida species was performed with vaginal-specimen wet mounts and KOH microscopy.

Data analysis. Data were double-entered by use of SPSS software (SPSS, Chicago) and were verified by comparison of line listings with the clinic files. Data analysis was performed with SPSS statistical software and S-PLUS (MathSoft, Seattle). For purposes of analysis, records were censored after the time of documented HIV-1 seroconversion. Visits were included in the analysis if they occurred within 3 months of the preceding visit. A seroconversion visit was defined as the first visit at which the HIV-1 serological test became positive.

The frequency of symptoms and signs during seroconversion and nonseroconversion visits was compared. ORs for the presence of symptoms and signs in seroconverters versus nonseroconverters were estimated by means of logistic regression models with generalized estimating equations, with adjustment for time since the last clinic visit because of the unequal intervals between visits. The use of generalized estimating equations adjusts for multiple visits by the same individual [7].

Symptoms and signs that were significantly associated with seroconversion visits in univariate analysis were entered in a multivariate logistic regression model, with HIV-1 status as outcome and with adjustment for time between clinic visits. Stepwise backward elimination was used to select the subset of signs and symptoms with P < .15. A scoring system was developed, with 1 point being assigned to each of these symptoms or signs. Sensitivity, specificity, and likelihood ratios were estimated from logistic regression modeling with generalized estimating equations. Confidence intervals for the likelihood ratios were calculated, as described by Simel et al. [8] and Dujardin et al. [9]. The positive and negative likelihood ratios represent how much more likely a test is to be positive or negative in diseased than in nondiseased individuals [9].

Previous SectionNext Section

Results

Study patients. From March 1993 through March 1998, 4007 women were screened for HIV-1 antibodies, and 53.5% were seropositive. One thousand fifty-three HIV-1—seronegative women were enrolled in the prospective cohort study, and 883 women returned for ≥1 follow-up visits. A total of 9851 follow-up clinic visits occurred over a 5-year period. The median follow-up time was 12 months, ranging from 1 to 60 months. One hundred sixty-two patients seroconverted during follow-up, yielding an annual HIV-1 infection incidence of 12% (162 sero-conversions per 1353 person-years).

Seven thousand seven hundred and thirty-five follow-up visits by 823 women occurred within 3 months of the previous clinic visit and were therefore eligible for analysis. This subset included 103 seroconversion visits. The median time between the last seronegative visit and the first seropositive visit was 1.1 months, ranging from 1 week to 3 months. For the participants included in this analysis, median age at enrollment was 26 years (range, 16–48 years), and only 1% of women were married. The median duration of education was 8 years (range, 0–18 years). Age at first sexual contact was 16 years (range, 7–28 years). At enrollment, the median duration of prostitution was 1 year (range, 0–24 years). The median frequency of sexual acts per week was 2 (range, 0–13), and the median number of different sex partners was 1 (range, 0–10). Twenty percent of the women smoked cigarettes, and 79% drank alcohol at least once per week.

Frequency of signs and symptoms. The frequency of clinical symptoms reported by the women at seroconversion visits was significantly greater than that reported at nonseroconversion visits (table 1). Fever was the most common symptom and was reported by 53% of seroconverting women and 27% of nonseroconverting women (OR, 2.8; 95% CI, 1.8–4.2). Other symptoms that were reported significantly more often at the seroconversion visit were vomiting (OR, 4.8; 95% CI, 2.9–8.1), diarrhea (OR, 3.1; 95% CI, 1.8–5.4), headache (OR, 2; 95% CI, 1.3–3), arthralgia (OR, 2.1; 95% CI, 1.3–3.4), myalgia (OR, 2.8; 95% CI, 1.6–4.7), skin rash (OR, 2.1; 95% CI, 1–4.2), and swollen lymph nodes (OR, 3.3; 95% CI, 1.5–7.5). There were trends toward an association between seroconversion visits and fatigue (OR, 1.6; 95% CI, 1–2.5) and pharyngitis (OR, 1.7; 95% CI, 1–3.1).

Table 1
View larger version:
Table 1

Clinical symptoms and signs in Kenyan women at the clinic visit when HIV-1—seroconversion was documented and at visits when they tested seronegative.

On physical examination, extrainguinal lymphadenopathy (OR, 6.1; 95% CI, 1.9–19.8), inguinal lymphadenopathy (OR, 9.5; 95% CI, 4.3–31.8), and vaginal candidiasis (OR, 2.7; 95% CI, 1.7–4.2) were observed more frequently at seroconversion visits than at nonseroconversion visits. There was also a trend toward higher frequency of cervical mucopus at seroconversion visits (OR, 2; 95% CI, 1–4.3). Oral thrush, oral hairy leukoplakia, conjunctivitis, maculopapular skin rash, and herpes zoster were seldom seen in this cohort.

Eighty-three (81%) of the 103 seroconverting women had ≥1 of the 11 symptoms or signs that were significantly associated with acute HIV-1 infection, and 53 (52%) had ≥3 of these symptoms or signs. For comparison, women at 3836 (52%) of 7323 nonseroconverting visits had ≥1 of these 11 symptoms or signs (P < .001), and 18% had ≥3 (P < .001). Among seroconverters, these 8 symptoms and 3 signs occurred with the following decreasing order of frequency: fever (53%), headache (44%), arthralgia (24%), myalgia (18%), vomiting (18%), diarrhea (17%), skin rash (9%), and swollen lymph nodes (7%); and vaginal candidiasis (27%), extrainguinal lymphadenopathy (3%), and inguinal lymphadenopathy (3%).

Acute HIV-1—infection symptoms were often of sufficient severity to prevent women from working. Forty-four percent of women at seroconversion visits reported having been too sick to work, compared with 15% at nonseroconversion visits (OR, 4; 95% CI, 2.7–6.1). Women at seroconversion visits also reported longer duration of work absence than did those at nonseroconversion visits. Seventeen percent of the women at seroconversion visits stayed home for >1 week, compared with 2% of the women at nonseroconversion visits (OR, 7.4; 95% CI, 4.1–13.1).

Sensitivity, specificity, and likelihood ratios for symptoms and signs. A scoring system was developed, as described in the Methods, that included 4 symptoms (fever, vomiting, diarrhea, and too sick to work) and 2 clinical signs (inguinal lymphade-nopathy and vaginal candidiasis). Sensitivity, specificity, and likelihood ratio were calculated, as shown in table 2. For a cutoff level of ≥2 symptoms or signs, the sensitivity for detecting primary HIV-1 infection was 51%, and the specificity was 83%. The positive likelihood ratio was 3.2, i.e., HIV-1—seroconverting women were ∼3 times more likely than non-seroconverting women to have ≥2 symptoms or signs. The corresponding negative likelihood ratio was 0.5, i.e., non-sero-converting women were twice as likely as seroconverting women to have a score <2.

Table 2
View larger version:
Table 2

Sensitivity, specificity, and likelihood ratios for different cutoff levels of scores for symptoms and signs.

Previous SectionNext Section

Discussion

In this prospective cohort study of female sex workers, we identified 11 symptoms and signs that were significantly associated with HIV-1 seroconversion. Four-fifths of seroconverting women had ≥1 of these symptoms or signs of primary HIV-1 infection. For 44% of the seroconverters, the illness was severe enough to prevent working, and 17% were out of work for >1 week. To our knowledge, this is the first characterization of the clinical manifestations of HIV-1 infection in an African population.

A major strength of this study was its prospective design. Monthly serological testing permitted identification of primary HIV-1 infections soon after the time of virus acquisition. Our estimate of the frequency of primary HIV-1 illness may therefore be more accurate than those in clinic- or hospital-based studies that identify seroconverting patients on the basis of presenting illness. A second related strength was the ability to compare seroconversion with nonseroconversion visits. This comparison revealed a relatively high prevalence of symptoms and signs in the seronegative control group, but illness in these women tended to be less severe. Only 15% of these women were too sick to work, and only a small fraction (2%) stayed home from work for >1 week.

We examined the usefulness of a checklist of selected symptoms and signs as a means of identifying people with recent HIV-1 infection. In any screening procedure, there is a trade-off between false-positives and false-negatives, since sensitivity and specificity are inversely related. At a higher positive likelihood ratio, seroconverters would be more likely to have symptoms and signs than nonseroconverters, but the sensitivity decreases, resulting in identification of fewer individuals with recent HIV-1 infection.

The choice of a cutoff point in the scoring system depends on the consequences and costs of false-positives relative to false-negatives and the prevalence of disease in the particular setting. This study population had a seroconversion-visit prevalence of 1.3%. If we selected a cutoff level of 2 in the checklist, sensitivity would be 51% and the specificity would be 83%, but the positive predictive value would be only 4%. In settings with a higher seroconversion-visit prevalence, the positive predictive value would increase. Because the sensitivity, specificity, and likelihood ratios are determined by the frequency of symptoms and signs during seropositive and seronegative visits, which may vary in different populations, our results cannot be extrapolated to all settings.

Recognition of primary HIV-1 infection is important for several reasons. First, the severity of symptomatic illness is of prognostic significance with respect to natural history of HIV-1 [10]. The plasma viral burden during acute HIV-1 infection or during the first year after seroconversion appears to establish the viral set point for the chronic phase of the disease [11, 12]. Second, the recognition of primary HIV-1 infection can prompt early initiation of antiretroviral therapy [13, 14]. Although clinical data are still limited, especially for long-term benefits, the Centers for Disease Control and Prevention and the International AIDS Society/USA Panel recommend that therapy should be initiated within 6 months of infection, before extensive immune system damage has occurred [15, 16].

Third, the study of primary HIV-1 infection can provide valuable information about the interaction of viral and host immunologic factors that may be of relevance for vaccine development [17]. Fourth, and most important from a public health perspective, identification and counseling of newly infected individuals may reduce transmission of HIV-1 during a period of increased infectivity. A substantial proportion of HIV-1 transmission overall is thought to be due to transmission from recently infected persons, because of transient high titers of cytopathic virus [2, 18].

Although the number of individuals with new HIV-1 infection is much smaller than the number of chronically infected individuals, the former contribute disproportionately to transmission [19]. Therefore, identification and counseling of recent seroconverters, with or without antiretroviral therapy, could have a marked effect upon the HIV-1 epidemic on a population level [20, 21].

In summary, in this cohort of women with high-risk behavior, HIV-1 seroconversion was accompanied by a high prevalence of symptoms and signs, consistent with acute HIV-1 illness. Although a number of the symptoms and signs were nonspecific, the occurrence of a constellation of characteristic clinical manifestations, particularly if of sufficient severity to prevent the patient from working, should alert a clinician to the possibility of primary HIV-1 infection and to offer HIV-1 testing.

Although antiretroviral therapy is not available for the majority of newly infected HIV-1 individuals in Africa, detection of primary HIV-1 infection could lead to risk-reduction counseling to prevent secondary transmission to sexual partners or to infants. In settings in which the HIV-1 epidemic is rapidly expanding and many primary infections are occurring, identification of newly infected individuals and appropriate counseling could be an important public health intervention.

Previous SectionNext Section

Acknowledgments

We thank Drs. Patrick Nyange (deceased), Salim Hassanali, Myra Maghasi, and Saade Abdalla; Claire Stevens; the staff of Ganjoni Clinic for their assistance with data collection; the Municipal Council of Mombasa for use of their facilities; and the women who participated in the study.

Previous SectionNext Section

Footnotes

  • Financial support: This project was supported by the National Institutes of Health through Family Health International (N01-AI-35173-119) and through grants AI-33873, D43-TW00007, and T22-TW00001.

  • Informed consent was obtained from all study participants, and the study was approved by the ethical review committees of the University of Washington and the University of Nairobi.

  • Received July 14, 1999.
  • Revision received October 29, 1999.
Previous Section
 

References

  1. ↵
    1. Kahn J,
    2. Walker B
    . Acute human immunodeficiency type 1 infection. N Engl J Med 1998;339:33-9.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Clark S,
    2. Saag M,
    3. Decker D,
    4. et al
    . High titers of cytopathic virus in plasma of patients with symptomatic primary HIV-1 infection. N Engl J Med 1991;324:954-60.
    MedlineWeb of Science
  3. ↵
    1. Quinn T
    . Acute primary HIV infection. JAMA 1997;278:58-62.
    Abstract/FREE Full Text
  4. ↵
    1. Fox R,
    2. Eldred L,
    3. Fuchs E,
    4. et al
    . Clinical manifestations of acute infection with human immunodeficiency virus in a cohort of gay men. AIDS 1987;1:35-8.
    MedlineWeb of Science
  5. ↵
    1. Bollinger R,
    2. Brookmeyer R,
    3. Mehendale S,
    4. et al
    . Risk factors and clinical presentation of acute primary HIV infection in India. JAMA 1997;278:2085-9.
    Abstract/FREE Full Text
  6. ↵
    1. Martin H,
    2. Nyange PM,
    3. Richardson B,
    4. et al
    . Hormonal contraception, sexually transmitted diseases, and risk of heterosexual transmission of HIV-1. J Infect Dis 1998;178:1053-9.
    Abstract/FREE Full Text
  7. ↵
    1. Zeger S,
    2. Liang K
    . Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121-30.
    CrossRefMedlineWeb of Science
  8. ↵
    1. Simel D,
    2. Samsa G,
    3. Matchar D
    . Likelihood ratios for continuous test results—making the clinicians' job easier or harder? J Clin Epidemiol 1993;46:85-93.
    CrossRefMedlineWeb of Science
  9. ↵
    1. Dujardin B,
    2. Van den Ende J,
    3. Van Gompel A,
    4. Unger J,
    5. Van der Stuyft P
    . Likelihood ratios: a real improvement for clinical decision making? Eur J Epidemiol 1994;10:29-36.
    CrossRefMedlineWeb of Science
  10. ↵
    1. Dorrucci M,
    2. Rezza G,
    3. Vlahov D,
    4. et al
    . Clinical characteristics and prognostic value of acute retroviral syndrome among injection drug users. AIDS 1995;9:597-604.
    MedlineWeb of Science
  11. ↵
    1. Mellors J,
    2. Kingsley L,
    3. Rinaldo C,
    4. et al
    . Quantitation of HIV-1 RNA in plasma predicts outcome after seroconversion. Ann Intern Med 1995;122:573-9.
    Abstract/FREE Full Text
  12. ↵
    1. Schacker T,
    2. Hughes J,
    3. Shea T,
    4. Coombs R,
    5. Corey L
    . Biological and virologic characteristics of primary HIV infection. Ann Intern Med 1998;128:613-20.
    Abstract/FREE Full Text
  13. ↵
    1. Volberding P,
    2. Lagakos S,
    3. Grimes J,
    4. et al
    . A comparison of immediate with deferred zidovudine therapy for asymptomatic HIV-infected adults with CD4 cell counts of 500 or more per cubic millimeter. N Engl J Med 1995;333:401-7.
    CrossRefMedlineWeb of Science
  14. ↵
    1. Kinloch-De Loes S,
    2. Hirschel B,
    3. Hoen B,
    4. et al
    . A controlled trial of zidovudine in primary human immunodeficiency virus infection. N Engl J Med 1995;333:408-13.
    CrossRefMedlineWeb of Science
  15. ↵
    Department of Health and Human Services and Henry J. Kaiser Foundation. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. MMWR Morb Mortal Wkly Rep 1998;47(RR-5):43-82.
  16. ↵
    1. Carpenter C,
    2. Fischl M,
    3. Hammer S
    . Antiretroviral therapy for HIV infection in 1997—updated recommendations of the International AIDS Society-USA Panel. JAMA 1997;277:1962-8.
    Abstract/FREE Full Text
  17. ↵
    1. Zhu T,
    2. Mo H,
    3. Wang N,
    4. et al
    . Genotypic and phenotypic characterization of HIV-1 in patients with primary infection. Science 1993;261:1179-81.
    Abstract/FREE Full Text
  18. ↵
    1. Daar E,
    2. Moudgil T,
    3. Meyer R,
    4. Ho D
    . Transient high levels of viremia in patients with primary human immunodeficiency virus type 1 infection. N Engl J Med 1991;324:961-4.
    MedlineWeb of Science
  19. ↵
    1. Leynaert B,
    2. Downs A,
    3. de Vincenzi I
    . Heterosexual transmission of human immunodeficiency virus. Am J Epidemiol 1998;148:88-96.
    Abstract/FREE Full Text
  20. ↵
    1. Cates W,
    2. Chesney M,
    3. Cohen M
    . Primary HIV infection—a public health opportunity. Am J Public Health 1997;87:1928-30.
    MedlineWeb of Science
  21. ↵
    1. Most J,
    2. Zangerle R
    . Absence of awareness of primary HIV infection. Lancet 1997;349:62.
    Medline
| Table of Contents

This Article

  1. Clin Infect Dis. (2000) 30 (3): 486-490. doi: 10.1086/313718
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most