Infection Pressure in Men Who Have Sex With Men and Their Suitability to Donate Blood

Abstract

Background

Deferral of men who have sex with men (MSM) from blood donation is highly debated. We therefore investigated their suitability to donate blood.

Methods

We compared the antibody prevalence of 10 sexually and transfusion-transmissible infections (TTIs) among 583 MSM and 583 age-matched repeat male blood donors. MSM were classified as low risk (lr) or medium-to-high risk (hr) based on self-reported sexual behavior and as qualified or unqualified using Dutch donor deferral criteria. Infection pressure (IP) was defined as the number of antibody-reactive infections, with class A infections (human immunodeficiency virus-1/2, hepatitis B virus, hepatitis C virus, human T-cell lymphotropic virus-1/2, syphilis) given double weight compared to class B infections (cytomegalovirus, herpes simplex virus-1/2, human herpesvirus 8, hepatitis E virus, parvovirus B19).

Results

Donors had a lower median IP than qualified lr-MSM and qualified hr-MSM (2 [interquartile range {IQR}, 1–2] vs 3 [IQR, 2–4]; P < .001). Low IP was found in 76% of donors, 39% of qualified lr-MSM, and 27% of qualified hr-MSM. The prevalence of class A infections did not differ between donors and qualified lr-MSM but was significantly higher in qualified hr-MSM and unqualified MSM. Recently acquired class A infections were detected in hr-MSM only. Compared to blood donors, human herpesviruses were more prevalent in all MSM groups (P < .001).

Conclusions

IP correlates with self-reported risk behavior among MSM. Although lr-MSM might form a low threat for blood safety with regard to class A infections, the high seroprevalence of human herpesviruses in lr-MSM warrants further investigation.

(See the Editorial Commentary by Haire and Kaldor on pages 1009–10.)

In response to the AIDS epidemic, lifetime donor deferral of men who have sex with men (MSM) was introduced to increase blood safety in the 1980s [1, 2]. The scientific rationale for MSM donor deferral has weakened over time, as nucleic acid amplification testing (NAAT) for human immunodeficiency virus (HIV) has substantially decreased the window period, during which HIV cannot yet be detected but may be transmitted via transfusion [3, 4]. As a result, public and political pressure to liberalize deferral policies has caused many high-income countries to change from permanent to temporary MSM donor deferral [1, 2, 5].

Currently, in Europe, donor deferral of MSM varies from “gender neutral” policies that are based on sexual practices without considering the gender of the sex partner (Spain, Italy, and Portugal), to temporary deferral of 3–12 months after last male-to-male sex contact (eg, United Kingdom, Germany, France, and the Netherlands), to permanent deferral (Denmark, Austria, and Croatia) [6]. In the Netherlands, MSM deferral is continued for sexually active MSM, defined as having male-to-male sexual contact in the past year. Despite a leveling off of the HIV incidence among Dutch MSM in recent years, MSM remain the major risk group for HIV infection in the Netherlands [7]. In 2016, 67% of the 816 new HIV diagnoses were in MSM who comprise an estimated 3% of the adult male population [7]. In addition, a recent phylogenetic study suggested that 24 of 32 (75%) Dutch and Flemish male donors diagnosed with HIV in the period 2005–2014 acquired their HIV infection through male-to-male sex [8].

To achieve a harmonized European approach on MSM donor deferral, the Council of Europe has requested epidemiological and behavioral data to support an evidence-based recommendation to both protect the blood supply and avoid stigmatizing population groups who are deferred unreasonably [9, 10]. High-quality studies investigating the risk of transfusion-transmissible infections (TTIs) in MSM who donate blood are scarce [11, 12]. In this study we investigated the suitability of MSM to donate blood by comparing the antibody reactivity to 10 sexually transmitted and/or blood-borne infections between MSM and the general male donor population. We hypothesized that MSM with self-reported low-risk behavior pose a low threat to blood safety if their infection pressure (IP) is comparable to that of male donors.

MATERIALS AND METHODS

Study Population

The Amsterdam Cohort Studies (ACS) among MSM is an open prospective cohort study initiated in 1984 to investigate the prevalence, incidence, and risk factors of HIV and other blood-borne and sexually transmitted infections [13]. Participants biannually visit the Public Health Service of Amsterdam, where they give blood for testing and storage and complete a standardized questionnaire about sexual behavior in the preceding 6 months. We selected all HIV-negative MSM with an ACS visit between January and August 2016. The ACS questionnaire was extended with donor eligibility questions from the Dutch Donor Health Questionnaire (DHQ) and questions regarding the willingness to donate blood.

For a primary comparison group, we randomly selected age-matched repeat male blood donors from blood collection sites in Amsterdam between January and August 2016. Because repeat donors went through several selection and testing cycles, a second comparison group included all men who registered as a (potential) new blood donors at a collection site in Amsterdam in that period.

Routine Donor Screening

Upon registration as blood donor and at every donation visit, donors complete a standardized DHQ, and their blood is screened for TTIs if no risk factors for donation are reported. New and repeat donors are routinely tested for antibodies to HIV types 1 and 2 (HIV-1/2), hepatitis B surface antigen (HBsAg), antibodies to hepatitis B core antigen (anti-HBc), antibodies to hepatitis C virus (HCV), and antibodies to human T-cell lymphotropic virus types 1 and 2 (HTLV-1/2; new donors only) using chemiluminescent immunoassays (PRISM, Abbott Laboratories), and for syphilis antibodies using particle agglutination test (TPHA PK 2000, Trinity Biotech). In addition, donors are tested for HIV-1/2 RNA, hepatitis B virus (HBV) DNA, and HCV RNA in pools of 6 donations (COBAS 6800 MPX assay, Roche Diagnostics). Confirmation testing includes alternative NAT for HIV-1/2, HBV, and HCV (Cobas AmpliPrep/Cobas TaqMan version 2.0, Roche Diagnostics); HBsAg neutralization (Architect HBsAg confirmatory test, Abbott); immunoblots for HIV-1/2, HCV, HTLV-1/2, and syphilis (INNO LIA Score, Fujirebio); and Venereal Disease Research Laboratory testing. Anti-HBc–reactive donors are tested for anti-HBs (Architect Anti-HBs, Abbott) and require an antibodies to hepatitis B surface antigen (anti-HBs) titer of 200 IU/mL for eligibility [14, 15].

Laboratory Methods

MSM and repeat donors were screened for antibodies against 10 sexually transmitted and/or blood-borne infections, categorized in classes A and B. Class A included the 5 infections that are part of routine donor screening (ie, HIV, HBV, HCV, HTLV, and syphilis). Class B infections included cytomegalovirus (CMV), herpes simplex virus 1 and 2 (HSV-1/2), human herpesvirus 8 (HHV-8), hepatitis E virus (HEV), and parvovirus B19. Class B infections were included as they are potential markers for (sexual) risk behavior for which the estimated prevalence in the general Dutch population varies from very low (HHV-8) to very high (parvovirus B19). Serological assays included Murex HIV antigen/antibody, CMV immunoglobulin G2, and HSV-1/2 immunoglobulin G (IgG) (Liaison, Diasorin); anti-HCV, anti-HBc II, HTLV-1/2, and Syphilis TP (Architect, Abbott Laboratories); Kaposi sarcoma–associated herpesvirus/HHV-8 IgG enzyme-linked immunosorbent assay kit (Advanced Biotechnologies); HEV IgG (Wantai Pharmacology Pharmacy Enterprise); and parvovirus B19 IgG (Biotrin). MSM found to be anti-HBc reactive were subsequently tested for HBsAg and anti-HBs (Architect, Abbott Laboratories).

Study Variables

Infection Pressure

The IP for each participant was defined as the number of antibody-reactive infections, with antibody-reactive class A infections arbitrarily given double weight. The cutoff to distinguish between low and high IP was set at the median IP determined in repeat male donors. If antibodies against any class A infection were detected, the IP was defined as high.

Recent Class A Infections

Recent class A infections were defined as antibody seroconversion within the preceding year. Antibody seroconversion was determined using previous test results and testing stored serum samples. This 1-year period was set because the current eligibility criteria for blood donation require 1 year without male-to-male sexual contact.

Sexual Risk Behavior

MSM were categorized as low risk (lr-MSM) or medium-to-high risk (hr-MSM) based on self-reported sexual behavior during the preceding year. Low-risk sexual behavior was defined as either (i) no anal intercourse, (ii) a monogamous relationship with a steady male partner (including condomless sex with that partner), or (iii) consistent condom use during anal intercourse with casual partner(s). All other MSM were categorized as hr-MSM. HIV preexposure prophylaxis was not included in MSM risk classification.

Qualification of MSM as Potential Blood Donors

All blood donors fulfilled the Dutch eligibility criteria for donation, as assessed in the DHQ. Currently, in the Netherlands, MSM face a temporary deferral of 12 months after the last male-to-male sexual contact. In this study, MSM were qualified as potential blood donors if they did not report any of the following Dutch permanent deferral criteria: a history of injecting drug use or commercial sex work, blood transfusion, or transplantation after 1980; a UK visit >6 months between 1980 and 1996; a family history of Creutzfeldt-Jakob disease; a previous diagnosis with HIV, HCV, HTLV, or syphilis; or an active HBV infection. Note that permanent deferral criteria in the Netherlands might differ from those used in other countries.

Willingness to Donate

For MSM, intention to donate was measured on a 7-point Likert scale (ranging from completely disagree to completely agree) using the following statement: “If I would be allowed to donate blood, I would apply as a blood donor in the future.” We distinguished moderate to high intention (score 4–7) from no to low intention (score 1–3).

Statistical Analysis

Age, IP, seroprevalence per infection, and number of recent class A infections of qualified lr-MSM, qualified hr-MSM, unqualified lr-MSM, and unqualified hr-MSM were compared to those of repeat donors using the unpaired t test (normal distributed numerical data), the Mann-Whitney U test (nonnormal distributed numerical data), or the χ² test/Fisher exact test (categorical data). MSM with missing data on risk behavior and/or eligibility were excluded from the main analysis. Additional analyses were performed to determine differences in the main outcomes within the group of qualified lr-MSM. A P value ≤ .05 was considered significant. Analysis was performed using Stata Intercooled version 13.1 software (StataCorp, College Station, Texas).

Sensitivity Analyses

Sensitivity analyses were performed to assess the impact of missing data. In addition, we repeated our analyses on IP using 2 alternative methods to calculate IP. First, class A infections were weighted 4 times more than class B infections. Second, IP was defined as the individual predictive probability of class A infections given all individual class B infections (Supplementary Data 2).

RESULTS

Study Population

During the study period, 583 of 604 (97%) MSM agreed to participate in our study. They were age-matched to 583 repeat male blood donors. We then excluded MSM with missing data on sexual risk behavior and/or eligibility, leaving 520 MSM for the main analysis (Figure 1). The median age of MSM and repeat donors was 42 years (interquartile range [IQR, 35–48 years for MSM and 34–48 years for donors). New donors had a median age of 28 years (IQR, 24–34).

Based on our sexual risk criteria, 245 of 520 (47%) were lr-MSM, and 275 of 520 (53%) were hr-MSM. Among lr-MSM, 52 of 245 (21%) reported no anal intercourse in the preceding year, 60 of 245 (24%) had a monogamous relationship, and 133 of 245 (54%) consistently used condoms during anal sex with casual partners. Based on our qualification criteria for blood donation, 380 of 520 (73%) MSM qualified as potential donors, and 140 of 520 (27%) MSM did not qualify. Of those unqualified, 92 reported a history of syphilis, 20 reported commercial sex work, 22 lived in the United Kingdom between 1980 and 1992, 9 had a history of transplantation, 7 received a blood product after 1980, 4 had ever injected drugs, 2 reported a family history of Creutzfeldt-Jakob disease, and 1 had a previous HCV diagnosis.

The majority of MSM had a moderate-to-high intention to donate (380/520 [73%]; median score, 5 [IQR, 4–6]). Intention did not differ among qualified lr-MSM, qualified hr-MSM, unqualified lr-MSM, and unqualified hr-MSM (P = .223).

Infection Pressure

The median IP among repeat donors (2 [IQR, 1–2]) was significantly lower than among qualified lr-MSM (3 [IQR, 2–4]), qualified hr-MSM (3 [IQR, 2–4]), unqualified lr-MSM (4 [IQR, 3–5]), and unqualified hr-MSM (4 [IQR, 4–6]) (P < .001 for all; Figure 2; Supplementary 1). The proportion of participants with a low IP (≤2) was 76% (443/583) in repeat donors, 39% (76/197) in qualified lr-MSM, 27% (50/183) in qualified hr-MSM, 13% (6/48) in unqualified lr-MSM, and 4% (4/92) in unqualified hr-MSM (P < .001 for all).

Class A Infections

None of the donors or MSM had HIV-1/2 antibodies. Anti-HCV was detected in 1 of 48 (2%) unqualified lr-MSM, 2 of 92 (2%) unqualified hr-MSM, and 1 of 361 (1%) new donors. Anti-HTLV-1/2 was detected in 1 of 92 (1%) unqualified hr-MSM. None of the repeat donors or qualified lr-MSM had antibodies against syphilis, while 2 of 361 (1%) new donors, 5 of 183 (3%) qualified hr-MSM, 23 of 48 (48%) unqualified lr-MSM, and 51 of 92 (55%) unqualified hr-MSM tested positive for syphilis. Anti-HBc reactivity was lowest in repeat donors (6/583 [1%]) and new donors (8/361 [2%]). It was significantly higher among MSM (P < .001), increasing from qualified lr-MSM (15/197 [8%]), qualified hr-MSM (17/183 [9%]), to unqualified lr-MSM (6/48 [13%]), to unqualified hr-MSM (26/92 [26%]). An anti-HBs titer <200 IU/L was found in 0 of 583 repeat donors, 4 of 361 (1%) new donors, 5 of 197 (3%) qualified lr-MSM, 8 of 183 (4%) qualified hr-MSM, 4 of 48 (8%) unqualified lr-MSM, and 9 of 92 (10%) unqualified hr-MSM. One qualified lr-MSM and 1 qualified hr-MSM had a chronic HBV infection (HBsAg-positive). We found no recent class A infections in repeat donors and (un)qualified lr-MSM. In qualified hr-MSM we found 2 recent syphilis infections and 2 recent anti-HBc seroconversions. In unqualified hr-MSM we found 6 recent syphilis infections and 1 recent anti-HBc seroconversion.

Class B Infections

Prevalence of antibody to HEV (anti-HEV) and to parvovirus B19 (anti–parvovirus B19) was similar among repeat donors and MSM, regardless of MSM group. It varied from 13% to 23% for anti-HEV and from 74% to 90% for anti–parvovirus B19. The seroprevalence of human herpesviruses (HHVs) was significantly higher among MSM compared to repeat donors (P < .001 for all). CMV antibody reactivity varied from 68% in qualified lr-MSM to 92% in unqualified hr-MSM, compared to 38% in repeat donors. HSV-1/2 antibody reactivity varied from 81% in qualified lr-MSM to 90% in unqualified hr-MSM, compared to 46% in repeat donors. HHV-8 antibody reactivity varied from 39% in qualified lr-MSM to 65% in unqualified lr-MSM, compared to 5% in repeat donors.

MSM With Low Sexual Risk Behavior

Additional analyses revealed no significant difference in seroprevalence of class A infections, HEV, or parvovirus B19 between 3 groups of qualified lr-MSM (Table 1). MSM who reported consistent condom use with casual partners had a higher prevalence of CMV, HSV, and HHV-8 (P < .001, P = .004, and P = .049, respectively), compared with MSM who reported no anal intercourse or a monogamous relationship in the preceding year.

Sensitivity Analysis

Results did not change when all 35 MSM with missing risk behavior data were classified as lr-MSM or hr-MSM. Likewise, results did not change when all 38 MSM with missing qualification data were classified as qualified or unqualified. Finally, our results remained similar if using 2 alternative approaches for IP calculation (Supplementary Data 2).

DISCUSSION

To our knowledge, this is the first study focusing on suitability of MSM to donate blood, taking into account self-reported sexual risk behavior linked to the seroprevalence of multiple sexually transmitted and TTIs. None of the qualified lr-MSM had acquired HIV, HBV, HCV, HTLV, or syphilis within the previous year and, except for HBV, the antibody prevalence of these 5 infections were comparable to prevalence in new and repeat male donors.

The anti-HBc prevalence among qualified lr-MSM was higher than in donors, but two-thirds of these MSM had anti-HBs titers >200 IU/L, making them eligible donors according to the current donor guidelines on HBV in the Netherlands [15]. One qualified lr-MSM who reported no male-to-male sex within the preceding year had a preexisting chronic HBV infection. This MSM complied with current donor eligibility criteria in the Netherlands, and his HBV infection would have been detected with routine donor screening. Qualified hr-MSM had an increased seroprevalence of both HBV and syphilis compared to donors and, importantly, 4 of 23 (17%) had acquired these infections within the last year. As recent infections may slip through routine donor screening as infectious “window-phase donations,” deferral of hr-MSM seems legitimate. In the Netherlands, universal vaccination of newborns without catch-up of adolescents was introduced in 2011 [16]. HBV vaccination is free of charge for MSM since 2002, and actively stimulated using a targeted vaccination program [17]. Over time, these preventive measures will result in a further decrease of HBV transmission among all citizens [17–19].

The median IP in qualified lr-MSM and hr-MSM was significantly higher than in repeat donors. For qualified lr-MSM, this was due to an increased seroprevalence of HHVs (ie, CMV, HSV, and HHV-8). Those who reported consistent condom use with casual partners had a significantly higher prevalence of HHVs than MSM in a monogamous relationship and MSM who had no anal intercourse in the preceding year. HHVs are easily transmitted by close contact, and could be a marker for increased sexual activity. Because CMV, HSV-1/2, and HHV-8 are cell associated, universal leukodepletion considerably reduces their risk of transmission via red blood cells and platelets. Their infectious burden in plasma for transfusion or fractionation is reduced using a combination of pathogen-inactivating and virus-reducing procedures. Although donors in the Netherlands are not screened for HHVs, it cannot be excluded that the increased seroprevalence of CMV, HSV-1/2, and HHV-8 observed in lr-MSM represents increased sexual risk taking that might compromise blood safety. Further research is needed on whether these and other potential laboratory markers are indeed relevant to maintaining blood safety.

The major strength of this study is that MSM were classified based on detailed information on sexual risk behavior collected in a prospective longitudinal cohort study. We used predefined criteria to distinguish between lr-MSM and hr-MSM, based on sexual behavior associated with acquiring HIV infection [20–22]. The criteria could easily be incorporated in the DHQ and are generally accepted and even suggested by MSM [23].

Our study has the following limitations. First, the study size might have resulted in a low power for accurate estimates and comparisons, in particular for low prevalent class A infections. Second, the infection pressure calculation is arbitrary. Class A infections were weighted double, because of their higher relevance to blood safety. However, alternative approaches to calculate IP did not substantially change our results. Third, ACS participants might not represent the larger MSM population living in the Netherlands. Due to its objectives and inclusion criteria, ACS enrolls sexually active MSM and might underrepresent MSM with very low sexual risk behavior. This selection bias might have led us to overestimate the IP among MSM. On the other hand, all MSM from the ACS were offered HBV vaccination. This might not be representative for the larger MSM population, leading to an underestimation of the IP. Fourth, antibody reactivity was used as a marker for lifetime sexual behavior and does not solely represent sexual behavior within the previous year. As sexual behavior fluctuates over life [24], the antibody reactivity in lr-MSM could result from behavior that has been discontinued. To partially correct for this, we distinguished between past and recent class A infections. Fifth, repeat donors do not represent the general population, as they are selected based on their low risk of TTIs. Dutch donors have a 6- to 60-fold lower TTI prevalence and incidence compared to the general population [14]. Moreover, repeat donors had been screened for HIV, HBV, HCV, and syphilis at every donation visit. If found positive, they would have been permanently deferred. Hence, by definition, class A infections among repeat donors would be recent infections among a selected group without a history of class A infections. New donors were included as a second comparison group. They have not previously been tested for class A infections and therefore are less likely to be subject to selection bias. Indeed the seroprevalence of HBV, HCV, and syphilis in new donors is higher than in repeat donors and closer to that of qualified lr-MSM.

CONCLUSIONS

Our study shows that MSM with low sexual risk behavior who did not report any of the criteria for permanent donor deferral in the Netherlands had no recent HIV, HBV, HCV, HTLV, or syphilis infections. Their seroprevalence of these infections was similar to new and repeat donors. Antibody prevalence to HHVs in lr-MSM was significantly higher than in repeat donors, but the epidemiological significance of this finding to recipient safety is unclear. Our results suggest that hr-MSM pose an increased threat to blood safety, as recently acquired HBV and syphilis infections were found in this group. Our results justify further study of the suitability of lr-MSM to donate blood, with additional focus on MSM compliance with guidelines and self-withdrawal, especially considering the high proportion of MSM with a moderate-to-high intention to donate.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. T. J. W. L., K. H., H. L. Z., and M. P. are responsible for the study design. M. F. S. L. and M. P. supervised data collection for MSM, and E. S. and T. J. W. L. supervised data collection for donors. T. J. W. L. was responsible for laboratory testing. W. P. H. B. performed the statistical analysis under supervision of A. M. and M. P.. All authors contributed to the interpretation of the results, writing of the manuscript, and providing intellectual feedback. All authors have seen and approved the final submitted version of the manuscript.

Acknowledgments. The authors gratefully acknowledge the Amsterdam Cohort Studies (ACS) on HIV Infection and AIDS, a collaboration between the Public Health Service of Amsterdam, the Academic Medical Center of the University of Amsterdam, the Sanquin Blood Supply Foundation, Medical Center Jan van Goyen, and the HIV Focus Center of the DC-Clinics. It is part of the Netherlands HIV Monitoring Foundation and financially supported by the Center for Infectious Disease Control of the Netherlands National Institute for Public Health and the Environment. The authors thank all ACS participants and blood donors for their contribution. On behalf of the authors, we also thank Alexandra Kovaleva and the research nurses of ACS for their contributions to data collection, Udi Davidovich for his valuable input on study design and interpretation of results, and Anders Boyd for his statistical advice and input.

Financial support. This study was financed by the Research and Development Foundation of the Public Health Service of Amsterdam and a Product and Process Development Cellular Products grant (PPOC 15-04) of Sanquin.

Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References

Author notes