Introduction

The human immunodeficiency virus (HIV) has had a dramatic effect on the incidence and effect of spondylodiskitis. By the end of 2015, 36.7 million people globally were living with HIV. During 2015, 2.1 million people had contracted the virus, with 1.1 million dying from it. This disease burden is concentrated in sub-Saharan Africa, which accounts for 25.6 million (70%) of the worldwide HIV-positive population. ¹ In this region of the world, a syndemic of HIV and tuberculosis (TB) exists, in which these two diseases converge synergistically to magnify the burden of disease. ² According to World Health Organization (WHO) data, there were 10.4 million new cases of TB in 2015, of which 1.2 million (11%) were HIV positive. Six countries—India, Indonesia, China, Nigeria, Pakistan, and South Africa—account for 60% of the TB cases. The highest TB/HIV co-infection was in southern Africa. In 2015, there were 1.4 million TB-related deaths, 0.4 million (29%) being HIV positive. The HIV incidence may well be underreported, as only 55% of TB patients worldwide had a documented HIV result, although the numbers are higher in southern Africa, where 81% of TB cases had a recorded HIV status. ³

The Cellular Impact of HIV and TB

Sexual contact among heterosexuals is the dominant mode of HIV transmission, although outside Africa, a third of cases are due to intravenous drug use. Mycobacterium tuberculosis is generally acquired via inhalation. The alveolar macrophages phagocytose the bacilli and process them in transit to the regional lymph node to initiate an adaptive immune response. The bacilli avoid fusion with lysosomes and continue to replicate. An exuberant accumulation of inflammatory cells form a multinucleated giant-cell–rich granuloma within which the bacilli persist. ⁴ Cell-mediated immunity is essential for control of M. tuberculosis infection. Activation of both CD4+ and CD8+ T cells are seen, and CD4+ T lymphocytes of T helper cell type 1(Th1) are thought to be most critical. ⁵

Human immunodeficiency virus gains access to cells without immediate lethal damage. It first binds to the CD4 receptor, triggering conformational changes that release the viral core into the cell cytoplasm. The viral genome is reverse transcribed into DNA by the virus’s own reverse transcriptase enzyme. This process is error prone, but distinct viral variants may be created. The viral genome is then inserted into gene-rich, transcriptionally active domains of the host’s chromosomal DNA, transforming the cell into a potential virus producer. The hallmark of HIV infection is the depletion of CD4+ T cells. ⁶ This leads to increased TB reactivation. One third of the world’s population is thought to be latently infected with M. tuberculosis, although the data supporting this notion may be questioned. Approximately 10% of M. tuberculosis–infected individuals are thought to develop overt clinical disease, and about half of them develop the disease longer than 2 years after infection. These cases are commonly termed reactivation or postprimary TB. The lifetime risk of developing active TB in immunocompetent adults is estimated to be 5 to 10%, but in HIV-positive individuals this risk is increased to 5 to 15% annually. ⁵

Not only does HIV increase the likelihood of TB reactivation, but also TB drives HIV replication by increased HIV-infected CD4+ cell reproduction. Phagocytosis of TB induces macrophage activation and proinflammatory cytokines [tumor necrosis factor-a (TNF-a), interleukin-1β (IL-1β), IL-6], which enhance HIV replication, driving the cycle further. ⁴

Clinical Presentation

The commonest general opportunistic infections in HIV+ patients reported from Uganda were candidiasis, diarrhea, geohelminths, TB, malaria, and pneumonia. The candidiasis was confirmed as commonest in an Indian study. ⁷ , ⁸

In the spine, M. tuberculosis remains the overwhelming co-infection. According to unpublished data I have gathered, 69 cases of spondylodiskitis were managed in the 2-year period of 2014–2015 at a single tertiary center base in Cape Town, South Africa; 31 of these cases were confirmed to be HIV positive, of which 26 were due to TB, four to pyogenic bacteria, and one to cryptococcal infection.

Not only does HIV increase the risk of contracting TB, but more importantly there is a 10% annual risk of reactivation as the contained M. tuberculosis bacillus is released. This is 12 to 20 times higher than in the immunocompetent individual. HIV also increases the incidence of extrapulmonary TB (EPTB), typically cited in immunocompetent patients as occurring in 10% of cases, half of which are musculoskeletal, the commonest being spinal TB at around 2% of total TB cases.

The clinical presentation is similar to the general presentation of spinal TB in the immunocompetent patient, with delayed presentation due to the insidious nature of the disease, the vague symptoms of pain and weight loss, and the lack of access to health care in the regions of the world where the HIV/TB syndemic exists.

Human immunodeficiency virus is reported as more common in men than in women, with a 2:1 ratio in the age range of 33 to 45 years. ⁹ This is reversed in southern Africa, where females predominate. This may well be due to woman having a higher risk of contracting HIV when exposed. TB may occur earlier in the HIV course than other opportunistic infections, due to the susceptibility of the M. tuberculosis–specific CD4+ cells to HIV and selective TB protective CD4-cell depletion.

Spinal TB may occur in the absence of pulmonary TB. By the time the patients present, there may be clinical kyphotic deformity (gibbus) or even a neurologic deficit. The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are usually raised, with the white cell count being normal.

A study conducted in a first-world hospital with 7,338 HIV-positive admissions in a 6-year period found that the incidence of spinal infection among admitted patients was calculated as 23.2 of 100,000 in the HIV-positive population compared with 7.1 of 100,000 in the HIV-negative population. The authors also found that of the 17 HIV-positive spine infections, eight were pyogenic with a CD4 count average of 339, six had TB with a CD4 count average of 75, and three had epidural abscesses with a CD4 count average of 21. ¹⁰

Diagnosis

Although imaging may be the same as in immunocompetent patients, that is, typically paradiskal kyphotic collapse with preservation of the disk space until later in the infection course, there may be less destruction seen on magnetic resonance imaging (MRI). With moderation of the cell-mediated type 4 hypersensitivity response, there may be less bony destruction and more pus collection. ¹¹ Early disk space height loss with intact bodies suggests pyogenic infection as opposed to the typical paradiskal erosions and kyphosis of M. tuberculosis spondylodiskitis.

Magnetic resonance imaging is useful to identify the pathology, determine the abscess size and location, confirm noncontiguous lesions, and assess the spinal cord. Up to 16% of TB spine patients may have noncontiguous vertebral lesions. ¹² MRI may confirm an HIV myelitis as a cause of neurologic deterioration rather than any spine infection ( Fig. 14.1 ).

Although TB remains by far the commonest infection in the HIV-positive cohort, other infectious or noninfectious causes must be considered, making spinal biopsy mandatory. A positive TB culture is the reference standard for identification of acid-fast bacilli. Microscopy may be negative due to the paucibacillary nature of the infection.

Increasingly, polymerase chain reaction (PCR) technology is utilized, which significantly reduces the 3- to 4-week period required to obtain a culture result. Xpert is a cartridge-based technology used at endemic sites for this very purpose and provides the result immediately. It amplifies a 81 base pair region of M. tuberculosis RNA polymerase ß-subunit (rpoß) gene. This is also the site of rifampin resistance, enabling confirmation of this drug’s usefulness. In spinal samples, Xpert has a sensitivity and specificity in the 95% range, although some cite it as being 10% lower in HIV-associated disease. This has not been our experience. As this is genetic material detection, it will be positive even after clinical cure, as it cannot distinguish between live and dead bacillus. ¹³