During a period when ART was routinely available in England, persons diagnosed with HIV had a 4.8-fold increased risk of IMD compared with individuals not known to be HIV infected. This increased risk was almost entirely in adults aged 16–64 years (23-fold increased risk) and was significant for Men C, W and Y, while MenB was only significant in those aged 16–24 years. There was only one MenB case in children with HIV and no cases in ≥65 year-olds. Most patients were born outside the UK, were known to be HIV-positive and receiving ART at the time of infection, and had not received any meningococcal vaccination prior to developing IMD.
These findings are consistent with the most recent study from New York City (NYC) where the relative risk for IMD among people living with HIV/AIDS (PLHIV) during the ART era was 10.0 especially among those with a CD4+ count of less than 200 cells/mm3 [13]. Notably, we observed a 15.9-fold increased risk for MenC, MenW and MenY, which are all potentially preventable through immunisation with the MenACWY quadrivalent conjugate vaccine [19]. The NYC study also found these capsular groups to be more prevalent among HIV-infected adults with IMD (87 %) compared to the IMD-only group (72 %), although the difference was not statistically significant. An earlier study from the United States also reported a significantly higher risk of IMD in HIV-positive adults [7] and these findings were corroborated in a large South African study, where the age-adjusted relative risk of developing IMD was 11.3, with a two-fold increased risk of dying from the infection [8]. Other studies from developing countries with high HIV prevalence, however, have not found any significant associations with IMD [9–12].
In HIV-uninfected individuals, half of all IMD cases in England are diagnosed in children younger than 5 years and more than 90 % of cases in this age group are caused by MenB [3]. Although only one IMD case was diagnosed among children with HIV, this single case represented a significantly higher risk for HIV-positive children when compared with HIV-uninfected children.
In adolescents, a small peak in IMD coincides with teenagers entering higher education settings (typically universities), where they are exposed to lifestyle factors and behaviors that increase their risk of IMD, such as smoking (or being exposed to smoke), kissing, sharing drinking glasses, eating utensils or water bottles, and living in close quarters (e.g. dormitories) [20]. Overall, MenB is still responsible for the majority (85–90 %) of IMD cases in adolescents. In HIV-positive persons, the increased risk in invasive MenB disease was only observed among 16–24 year-olds. In contrast, the risk of MenW and MenY disease was significantly higher across all adult age groups. These two capsular groups are uncommon causes of IMD in England and generally considered to be less virulent than MenB or MenC, causing disease mainly in older adults who often have underlying co-morbidities [4, 21]. A notable feature of these two capsular groups is the atypical clinical presentation, including respiratory tract infections such as pneumonia, epiglottitis, cellulitis and septic arthritis [22]. If IMD is not considered in the differential diagnosis, then public health actions to offer rapid chemoprophylaxis and the MenACWY conjugate vaccine to close contacts (who may also be HIV-positive) could be delayed and lead to secondary cases.
Although we observed significantly higher rates of capsular group-specific IMD in HIV-positive individuals, there were only 14 cases diagnosed over the three-year period, equivalent to 0.5 % of all IMD cases. In this high-risk group, however, we found that most individuals had acquired HIV through heterosexual contact and the incidence in MSM was similar to the background rate. A recent MenC disease outbreak in NYC was reported among MSM and subsequently identified in other countries, including Germany and France, resulting in an offer of vaccination against MenC to MSM in these countries [13, 23–28]. In England, there were no cases of MenC disease in children, adolescents or young adults (<25 year-olds) with HIV over the three-year period and none among MSM of any age, highlighting the success of the current national MenC conjugate vaccination programme in providing both direct and indirect protection against MenC disease [29].
Unlike the US, MenB is responsible for most IMD cases in children and adults in England. The significantly higher risk of MenC, MenW and MenY disease among persons diagnosed with HIV in England is, therefore, intriguing. Most cases were known to be HIV-positive at IMD diagnosis and half of them continued to have low CD4 counts at IMD diagnosis, making them generally susceptible to serious bacterial infections, including the less virulent meningococcal capsular groups such as MenW and MenY. In persons known to be HIV-positive, we found no association between IMD and CD4 counts at HIV diagnosis, time since HIV diagnosis, CD4 counts at IMD diagnosis or receiving ART at IMD diagnosis. Given that most cases had been born outside the UK, it is likely that they were at higher risk of exposure to infection when they themselves, their close family or community contacts travelled to their country of origin. It is also possible that these cases socialise in highly-defined closed networks – either because of their HIV infection or because of their ethnic background – where the less common meningococcal capsular groups may be circulating.
Limitations
Until now, except for complement deficiency and splenic dysfunction, HIV or any other immune deficiency has not been considered a significant risk factor for IMD [5]. In the current study, too, only 0.5 % of IMD cases were known to be HIV-positive. However, the IMD surveillance questionnaires were completed by GPs, who are not always aware of the diagnosis of HIV. It is, therefore, possible that HIV prevalence among IMD patients may be underestimated as demonstrated by the additional ascertainment through linkage with the national HIV database. In children, however, there were limited identifiers in the HIV database for linkage compared to adults and, therefore, some children with HIV and IMD may have been missed. It is also not possible to estimate the number of undiagnosed HIV-positive cases in the IMD cohort because the latter are not routinely tested for HIV. Furthermore, it is estimated that a quarter of persons estimated to be living with HIV are unaware of their infection and would not be included within the national cohort of persons accessing HIV care. However, the coverage of persons diagnosed and utilising HIV-related care indicates high rates of access to care and low rates of loss to follow-up [30]. Moreover, identifying additional HIV-positive individuals among IMD cases would only serve to strengthen the association between HIV and IMD, and increase the estimated relative risks further.
Clinical implications
A protein-based, multi-component MenB vaccine has recently been licensed in Europe and will be introduced into the UK infant immunisation programme in 2015 [31]. This vaccine induces high concentrations of bactericidal antibodies against most MenB strains causing IMD in the UK and could potentially also protect against other meningococcal capsular groups [32]. HIV-infected infants born in the UK will, therefore, benefit from the vaccine programme, although there are currently very few infants infected through MTCT because of the high uptake of antenatal screening and very high effectiveness of antenatal and perinatal ART in preventing MTCT from mothers known to be HIV-positive [33]. Older children with HIV, including newly diagnosed entrants to the UK, would not be eligible for MenB vaccination because there is no catch-up programme planned for the UK. Reassuringly, though, invasive MenB disease is much less common in older children irrespective of HIV status.
In adolescents and young adults, the recent increase in invasive MenW disease in England caused by a hypervirulent clonal complex (cc11) clone has led to recommendations for a national adolescent MenACWY conjugate vaccination programme for 14–18 year-olds, which will be introduced imminently [34]. This programme should, therefore, also protect HIV-positive individuals of all ages against these four capsular groups through both direct and indirect (herd) protection. The potential high level of exposure to meningococcal for infected individuals born overseas, however, may be less amenable to prevention by a UK-based programme and may reflect a substantially higher risk in uninfected people with overseas origins.
The MenACWY conjugate vaccine, however, will not protect against MenB disease. Future studies on the immunogenicity of the recently licensed MenB vaccine in HIV-positive children and adolescents at different stages of immunosuppression and antiretroviral treatment could potentially support recommendations for MenB vaccination in these age groups. Our results also indicate that newly diagnosed adults with HIV may benefit from MenACWY conjugate vaccination [19, 35]. However, the wide range of CD4 counts, including a significant proportion with very low counts, and the variable duration and combinations of ART at the time of IMD diagnosis suggest that these parameters may not be useful criteria for recommending the most appropriate timing for meningococcal vaccination. Current UK guidelines suggest that vaccination should be given to persons with CD4 counts <200 cells/mm3 if indicated and safe, and repeated following immunoreconstitution if required [36].
There are three available MenACWY vaccines, each conjugated to a different carrier protein, and one (MenACWY-D; Menactra®, Sanofi Pasteur, Lyon, France) has been assessed in HIV-positive children and adolescents; the vaccine was immunogenic against the four capsular groups and a two-dose elicited higher bactericidal antibody concentrations and protected more individuals than a single dose [19, 37]. Vaccine responses were dependent on age, the presence of an AIDS-defining illness, CD4 count and HIV viral load [19, 35, 37], although none of these factors were consistently predictive of poorer antibody responses. Unlike the UK, a two-dose MenACWY primary MenACWY conjugate vaccine schedule, given at least 2 months apart, is recommended for HIV-positive adolescents (>11 years) in the US, with boosters every 5 years [38]; notably, the recently licensed MenB vaccines have been recommended for those considered at increased risk of MenB disease but not for HIV-positive individuals [39].
Another important observation was that HIV diagnosis was known at the time of IMD for nearly all cases. In only two patients was the diagnosis of HIV made at the time of IMD infection and, in one case, after recovering from IMD, highlighting the importance of considering an underlying diagnosis of HIV if clinically indicated, including but not limited to immigration from an HIV-endemic area, intravenous drug use or MSM, history of opportunistic infections or low lymphocyte counts when hospitalised for IMD.
Availability of data and materials
All data for persons co-infected with HIV and invasive meningococcal disease are available within the Table 4. Further requests regarding individual participant data should be made to the clinical leads, Dr Shamez Ladhani for meningococcal (shamez.ladhani@phe.gov.uk) and Dr Valerie Delpech for HIV (valerie.delpech@phe.gov.uk).
