Pyogenic vertebral osteomyelitis is the most commonly encountered form of vertebral infection. It can develop from direct open spinal trauma, from infections in adjacent structures, from hematogenous spread of bacteria to a vertebra, or postoperatively. Left untreated, vertebral osteomyelitis can lead to permanent neurologic deficits, significant spinal deformity, or death. [1, 2, 3, 4, 5, 6, 7] It can result in severe compression of the neural stuctures due to formation of an epidural abscess or due to a pathologic fracture resulting from bone softening.
Evidence of vertebral osteomyelitis has been found in Egyptian mummies. Hippocrates first described the infection of the vertebral column. Later on, Galen related this infectious process to spinal deformity. Before the development of antibiotics and bacteriology, little knowledge was added to the basic understandings of the Hippocratic school until Servino and Potts characterized and described the pathology of tuberculosis infection of the spine. In 1879, Lannelonge described bacterial osteomyelitis as we recognize it today.
Although successful treatment of spinal abscess with surgical drainage was reported early on, the high complication rate from secondary infection caused this surgery to remain in poor favor. Following the introduction of antisepsis, surgical intervention for spinal infections became feasible.
The initial procedure introduced for the surgical treatment of spinal infections was a laminectomy. However, this procedure did not allow access to anterior abscesses and contributed to spinal instability, which often resulted in progressive deformity. Ito et al first described the anterior approach to the spine. Later, Hodgson and Stock extensively reported this procedure in the treatment of tuberculosis of the spine. Late spinal deformity was prevented with spinal fusion and instrumentation. While Hodgson and Stock performed fusions from the anterior approach, Hibbs and Albee independently presented techniques for posterior spinal fusion in the treatment of tuberculosis of the spine.
In the future, the introduction of newer, more effective antibiotics may contribute to the treatment of these infections. For patients requiring a fusion, the use of growth factors for the induction of spinal fusions is a theoretically attractive approach. Numerous studies have shown that viral vectors can be used to implant osteoinductive growth factor genes directly into the paraspinal muscles or into cells that can subsequently be implanted next to the spine. These osteoinductive factors enhance the activation and differentiation of pluripotent stem cells to develop into mature bone.
The anatomy of the spine includes the vertebral bodies, intervertebral disks, and associated joints, muscles, tendons, ligaments, and neural elements.
The intervertebral disk is a fibrocartilaginous remnant of the embryonic notochord, which provides the spine with strength, mobility, and resistance to strain. It consists of the following three parts:
The annulus fibrosus is made up of type I collagen fibrils, which are arranged in 15-20 concentric lamellae brought together into parallel bundles. These bundles are firmly attached to the vertebral bodies and are arranged in layers to provide strength and limit vertebral movement when the disk is compressed. The nucleus pulposus is composed of type II collagen and represents 30-60% of the disk volume. The nucleus pulposus is supplied with blood vessels through small perforations in the central cartilaginous endplates.
The cervical spine consists of the first seven vertebrae in the spinal column (C1-7). Typically, these vertebrae are small and possess a foramen on the transverse process for the vertebral artery. The thoracic spine consists of the next 12 vertebrae (T1-12) and is stabilized by the attached rib cage and intercostal musculature. The lumbar spine consists of a mobile segment of five vertebrae (L1-5), located between the relatively immobile segments of the thoracic and sacral segments.
The lumbar vertebrae are particularly large and heavy in comparison with the cervical and thoracic vertebrae. The bodies are wider and have shorter and heavier pedicles, and the transverse processes project somewhat more laterally and ventrally than the other spinal segments. The laminae are shorter vertically than the bodies and are bridged by strong ligaments. The spinal processes are broader and stronger than those in the thoracic and cervical spine.
Approximately 95% of pyogenic spinal infections involve the vertebral body, and only 5% involve the posterior elements of the spine. This disparity has been attributed in part to the voluminous blood supply to the vertebral body and its rich, cellular marrow.
Bacteria circulating through the blood may enter a vertebra or a disk space via its arterial blood supply or via the venous system. In the typical case, bacteria enter the vertebral body through small metaphyseal arteries arising from larger primary periosteal arteries that, in turn, branch from the spinal arteries. In adults, blockage of metaphyseal arteries by septic thrombi may infarct relatively large amounts of bone. Subsequently, bacteria can readily colonize a large bony sequestrum adjacent to the disk.
In the adult, after bacterial colonization of the metaphyseal region, the avascular disk is secondarily invaded by bacteria from the endplate region. Intermetaphyseal communicating arteries allow the spread of septic thrombi from one metaphysis to the other in a single vertebral body without involvement of the midportion of the vertebra.
Although the arterial route is the usual route of bacterial spread to a vertebra, another proposed route of infection is the retrograde seeding of venous blood via the Batson plexus. During periods of increased intra-abdominal pressure, venous blood is shunted toward the vertebral venous plexus. Some authors have proposed that the venous system may be the route of bacterial spread from genitourinary tract infections.
Another possible means of infection is by the spread of contiguous infection into the vertebrae and disk (eg, from a retropharyngeal abscess or a retroperitoneal abscess), resulting in osteomyelitis and diskitis. 
Presumably, a distant focus of infection provides an infective nidus from which bacteria spread by the bloodstream to the spinal column. The skin and the genitourinary tract are common antecedent sites, but a review of the literature reveals multiple foci, such as septic arthritis, sinusitis, subacute bacterial endocarditis, and respiratory, oral, or gastrointestinal infection. [9, 10, 11, 12] Approximately 30-70% of patients with vertebral osteomyelitis have no obvious prior infection.
Risk factors for developing osteomyelitis include conditions that compromise the immune system, such as the following:
Advanced age 
Long-term systemic administration of steroids
Diabetes mellitus 
IV drug abuse is a growing cause of spinal infections. Typically, the organism most likely to infect the spine is Staphylococcus aureus; however, in IV drug users, Pseudomonas species are also a common cause.  Nonpyogenic osteomyelitis can be caused by tuberculosis, fungus, yeast, or parasitic organisms. [17, 18, 19, 20, 21]
Surgical site infection (SSI) can result as an adverse event after a spinal procedure. Timing of preoperative antibiotic prophylaxis as well as careful aseptic technique can reduce the incidence of SSIs during spinal procedures. [22, 23]
Fungal infections of the spine are rare and generally occur in patients who are debilitated or have diabetes or a compromised immune system. Patients with acute leukemia, alcoholics, patients with lymphoma, recipients of organ transplants, and those receiving chemotherapy are particularly susceptible to fungal infections.
Most vertebral body infections occur in the lumbar spine because of the blood flow to this region of the spine. Tuberculosis infections have a predilection for the thoracic spine, and IV drug abusers are more likely to contract an infection of the cervical spine.
Vertebral osteomyelitis is considered uncommon, with an incidence of 1 case per 100,000-250,000 population per year. However, some reviews suggest that the incidence of spinal infections is now increasing. This increase may be secondary to increased use of vascular devices and other forms of instrumentation and to increasing rates of IV drug abuse.  Because of its rarity and vague initial signs and symptoms, diagnosis is often delayed.
No specific predilection for a particular race has been noted. Osteomyelitis has a predilection for males. A bimodal age distribution occurs in diskitis. Diskitis and osteomyelitis peak in pediatric patients; the incidence of spinal infections then decreases until middle age, when a second peak in incidence is observed at approximately age 50 years.  Some authors argue that childhood diskitis is a separate disease entity and should be considered independently.
In developed nations, the incidence of spinal osteomyelitis is similar to that in the United States. However, in less developed nations, infectious osteomyelitis is more common. In some areas of Africa, a reported 11% of all patients seen for back pain were diagnosed with diskitis and osteomyelitis.
Both bony and neural status must be considered in the evaluation of treatment outcome.  Most patients can be cured by a treatment protocol that includes antibiotics alone or in combination with surgery. [27, 28] For patients with an incomplete neurologic compromise, several studies indicate that with aggressive antibiotic and surgical therapy, paresis may improve or resolve. [29, 30, 31] Only 15% of patients experience permanent neurologic deficits. Recrudescence of infection occurs in 2-8% of patients.
In a retrospective cohort study, Gupta et al assessed 260 patients with pyogenic vertebral osteomyelitis, of whom 27% acquired the infection after an invasive spinal procedure, 40% had S aureus as the cause of the infection, and 49% underwent spinal surgery as part of initial therapy.  The estimated cumulative probability of treatment failure-free survival was 72% at 2 years, 69% at 5 years, and 69% at 10 years. On multivariate analysis, the factors associated with greater likelihood of treatment failure were (1) a longer duration of symptoms before diagnosis and (2) an infection caused by S aureus.
What would you like to print?