Bauer, Martijn P. MD*; Marijt, Erik W. A. MD, PhD†; Kroon, Frank P. MD, PhD*
Departments of *Infectious Diseases; and
†Hematology, Leiden University Medical Center, Leiden, The Netherlands
The authors have no funding or conflicts of interest to disclose.
To the Editors:
HIV infection is associated with hyposplenism.1–3 Hyposplenism is characterized by not only decreased splenic macrophage filter function but also decreased humoral immune responses, especially against T-cell independent antigens, such as pneumococcal polysaccharides,4 and may contribute to humoral immunodeficiency found in HIV-infected patients.5 Thus, hyposplenism may contribute to the increased risk of invasive pneumococcal disease in HIV-infected patients.6,7 However, little is known about the dynamics and predictors of hyposplenism in HIV infection. We sought to confirm the association of HIV infection with hyposplenism and investigate predictors of hyposplenism in HIV-infected patients, using the databases of the hematological laboratory and the infectious disease outpatient clinic of Leiden University Medical Center. Howell-Jolly bodies (HJBs), nuclear remnants in red blood cells normally cleared by the spleen, in peripheral blood smears were used as a marker for hyposplenism.8 Laboratory personnel, who are unaware of a patient’s HIV infection, always report HJB in smears.
Using the hematological database, we assessed the presence of HJB in all smears from HIV-infected patients between March 1995 and July 2007. As a control group, we examined all smears from non–HIV-infected patients in January 2005. HIV-infected patients were excluded if they had competing causes of hyposplenism, such as sickle cell anemia, which was the case in 1 splenectomized patient.
A total of 3993 smears were made in 365 HIV-infected patients. HJBs were seen in 61 smears (1.5% of all smears) from 36 patients (9.9% of all patients). A total of 1716 smears were made in 1050 control patients. HJBs were seen in 8 smears (0.5% of all smears) from 8 patients (0.8% of all patients). Using smears rather than patients to assess the association between HIV infection and HJB, we calculated an odds ratio (OR) of 3.31 [95% confidence interval (95% CI): 1.58 to 6.94].
To assess HJB persistence, 1 follow-up smear was made in each of 29 consenting HIV-infected patients from September 2007 to November 2008. HJBs were present in the smears of 7 of these patients and absent in the smears of the remaining 22 patients.
To investigate determinants of hyposplenism in HIV infection, we composed a control group of blood smears without HJB by random selection from all available smears from HIV-infected patients, matching for date to minimize the effect of historical changes in HIV management. This resulted in a database consisting of 155 smears from 91 HIV-infected patients: 68 smears with HJB from 36 patients and 87 smears without HJB from 78 patients. For each smear, we assessed the variables given in Table 1, in addition to symptomatic or latent syphilis, active tuberculosis, or infection with Mycobacterium avium intracellulare complex, Kaposi sarcoma, and lymphoma, by reviewing patient files. For continuous variables, we used the value measured at the date closest to the date of the smear with a maximum of 6 months before or after the date of the smear. The binary variables were scored positive if present at the date of the smear. Hepatitis B was defined as positive HBsAg, hepatitis C as positive hepatitis C RNA in plasma and symptomatic or latent syphilis as Treponema pallidum antibodies and positive Venereal Diseases Research Laboratory test.
Age was the only normally distributed variable as tested by Kolmogorov–Smirnov test. Mean ages in both groups were compared with a t test. The distributions of the other continuous variables were compared with a Mann–Whitney U test. For the association of binary variables with HJB, ORs with 95% CIs were calculated. For all statistical tests, computer software (SPSS Statistics version 20; IBM Corporation, Somers, NY) was used. Syphilis was found in 2 patients with HJB, tuberculosis in 4 patients with HJB and 1 patient without HJB, M. avium intracellulare complex infection in 1 patient without HJB, Kaposi sarcoma in 5 patients with HJB and 1 patient without HJB, and lymphoma in 1 patient with HJB and 3 patients without HJB. The other results are shown in Table 1. Generally, patients with HJB had more-symptomatic HIV infection, higher viral load, lower CD4 and CD8 counts, and shorter duration of combination antiretroviral therapy (cART). For continuous variables, receiver-operating characteristics for the prediction of HJB were made and optimal cutoffs were determined. CD4 count had an area under the receiver-operating characteristics curve of 0.601, CD8 count of 0.642, HIV viral load of 0.580, and cumulative duration of cART in patients with treatment indication of 0.659. The optimal cutoffs were 325 cells per microliter for CD4 count, 870 cells per microliter for CD8 count, detectable (versus nondetectable) for viral load, and 300 days for cumulative duration of cART in patients with treatment indication. CD4 cell count <325 cells per microliter, CD8 cell count <870 cells per microliter, CDC class C (AIDS-defining conditions),9 and cART duration under 300 days were strong predictors of HJB. A logistic regression model with AIDS-defining condition, CD4 cell count <325 cells per microliter, and cumulative cART duration <300 days in patients with treatment indication as covariates showed CD4 cell count <325 cells per microliter to be the strongest predictor of HJB (adjusted OR: 3.62; 95% CI: 1.45 to 9.04), whereas the other predictors were weakened.
We have shown that HIV-associated hyposplenism is associated with clinical disease stage and may be reversed by successful cART. The main limitations of this study are the retrospective design and the nonquantitative and relatively insensitive test for hyposplenism.8
Postmortem splenic histology in patients with AIDS shows prominent germinal center and marginal zone atrophy, whereas the mantle zone remains intact.10,11 Furthermore, splenic marginal zone type IgM memory B cells are preferentially depleted from the blood in HIV infection.12,13 Two studies found that blood IgM memory B cell count correlates with CD4 cell count,12,14 whereas 1 study did not.13 Postmortem histology of splenic red pulp in patients with AIDS shows macrophage dysfunction.10 Therefore, we believe hyposplenism, leading to decreased splenic macrophage filter function and decreased IgM response against polysaccharide antigens, contributes to HIV-induced immunodeficiency. Demonstrating hyposplenism may identify those HIV-infected patients with the highest risk of HIV-related infections and, thus, contribute to the decision to start cART.
The authors thank Theo de Jong and Henk Baelde for their detailed information on the examination of blood smears and help in data retrieval, the hematological laboratory for making and assessing the blood smears, and Juul Noomen and Willemien Dorama for their help with acquiring follow-up blood smears.
1. Bender BS, Frank MM, Lawley TJ, et al.. Defective reticuloendothelial system Fc-receptor function in patients with acquired immunodeficiency syndrome. J Infect Dis. 1985;152:409–412.
2. Corazza GR, Zoli G, Ginaldi L, et al.. Tuftsin deficiency in AIDS. Lancet. 1991;337:12–13.
3. Grotto HZ, Costa FF. Hyposplenism in AIDS. AIDS. 1991;5:1538–1540.
4. Kruetzmann S, Rosado MM, Weber H, et al.. Human immunoglobulin M memory B cells controlling Streptococcus pneumoniae
infections are generated in the spleen. J Exp Med. 2003;197:939–945.
5. Kroon FP, van Dissel JT, Ravensbergen E, et al.. Antibodies against pneumococcal polysaccharides after vaccination in HIV-infected individuals: 5-year follow-up of antibody concentrations. Vaccine. 1999;18:524–530.
6. Arthur G, Nduba VN, Kariuki SM, et al.. Trends in bloodstream infections among human immunodeficiency virus-infected adults admitted to a hospital in Nairobi, Kenya, during the last decade. Clin Infect Dis. 2001;33:248–256.
7. Kyaw MH, Rose CE Jr, Fry AM, et al.. The influence of chronic illnesses on the incidence of invasive pneumococcal disease in adults. J Infect Dis. 2005;192:377–386.
8. Di SA, Carsetti R, Corazza GR. Post-splenectomy and hyposplenic states. Lancet. 2011;378:86–97.
9. Castro KG, Ward JW, Slutsker L, et al.. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Recomm Rep. 1992;41:1–19.
10. Falk S, Stutte HJ. The spleen in HIV infection–morphological evidence of HIV-associated macrophage dysfunction. Res Virol. 1990;141:161–169.
11. Wilkins BS, Davis Z, Lucas SB, et al.. Splenic marginal zone atrophy and progressive CD8+ T-cell lymphocytosis in HIV infection: a study of adult post-mortem spleens from Cote d’Ivoire. Histopathology. 2003;42:173–185.
12. Hart M, Steel A, Clark SA, et al.. Loss of discrete memory B cell subsets is associated with impaired immunization responses in HIV-1 infection and may be a risk factor for invasive pneumococcal disease. J Immunol. 2007;178:8212–8220.
13. Morrow M, Valentin A, Little R, et al.. A splenic marginal zone-like peripheral blood CD27+. AIDS Res Hum Retroviruses. 2008;24:621–633.
14. D’Orsogna LJ, Krueger RG, McKinnon EJ, et al.. Circulating memory B-cell subpopulations are affected differently by HIV infection and antiretroviral therapy. AIDS. 2007;21:1747–1752.