Infections in Patients With Spinal Cord Injury

Factors That Predispose to Infection

Risks for infection in patients with SCI can be either systemic or organ related. As SCI does not, in and by itself, depress general host immunity, uninfected individuals with SCI usually have normal function of T and B lymphocytes. However, elevated levels of circulating proinflammatory cytokines and autoantibodies are present in the serum of patients with SCI without medical complications and are further elevated in such patients with neuropathic pain, UTI, or pressure ulcers relative to healthy, able-bodied individuals. Compared with able-bodied cohorts, patients with SCI, particularly individuals with tetraplegia, usually have higher levels of inflammatory markers such as C-reactive protein and cytokines (e.g., interleukin-6 and tumor necrosis factor-α), and this difference can be attributed to undetected inflammation or occult infection. Receipt of high-dose glucocorticosteroids in the acute setting of SCI and chronic complicating conditions (including stress, malnutrition, and renal failure) can impair the immune responses to infection. An animal model showed decreased local inflammatory markers in the bladder tissue of mice with SCI during infection and a delayed antiinflammatory response after being treated, which may contribute to the increased risk of UTI and the chronic inflammation that can persist after treatment.

More importantly, patients with SCI possess unique organ-related factors that predispose them to infection. For instance, most patients have a neurogenic bladder and have frequent episodes of UTI attributed to urinary stasis and bladder catheterization. Urinary stasis greatly impairs the naturally protective mechanisms of the urinary tract, including the washout effect of voiding and phagocytic capacity of bladder epithelial cells. Although some techniques of bladder catheterization are safer than others, none can be carried out without the potential risk for introducing organisms into the urinary tract. Both paralytic ileus and abnormal state of consciousness caused by associated head injury or illicit drug ingestion or both can predispose to aspiration pneumonia in the acute stage of SCI. In individuals with cervical or high thoracic cord lesions, weakness of the diaphragmatic and intercostal muscles impairs the capacity to clear respiratory secretions. Skin breakdown in anesthetic areas, immobility, disuse-induced muscle atrophy, urinary leakage, and fecal contamination predispose to infection of pressure sores.

Frequent insertion of urologic, vascular, orthopedic, respiratory, gastrointestinal, and neurosurgical devices in patients with SCI predisposes to various prosthesis-related infections. This helps explain why hospital-acquired infections most commonly affect the urinary tract, bloodstream, and musculoskeletal system. Patients with SCI generally have a higher rate of hospital-acquired infections than other groups of patients. About one-third of patients with SCI develop infection during hospitalization, with an overall incidence of 35 episodes of hospital-acquired infections per 1000 hospital-days.

Challenges in Evaluating Patients for Infection

A number of unique challenges can be encountered when attempting to establish diagnosis and provide treatment of infections in patients with SCI ( Table 309.1 ). Infection often manifests differently in patients with SCI compared with able-bodied individuals. Altered or absent sensations are the most important impediment to the diagnosis of infection. Although dysuria, frequency, and urgency are regularly present in able-bodied patients with UTI, these symptoms rarely exist in infected patients with SCI. The diagnosis of perinephric abscess is particularly challenging in patients with high sensory level lesions who do not appreciate flank pain or tenderness. The inability to recognize the signs and symptoms of cord damage contributes to the delay in diagnosing spinal epidural abscess below the level of injury. The diagnostic dilemma caused by the paucity of clinical findings can be heightened by the presence of neurogenic or referred pain that may not be related to the infection. Furthermore, multiple infections can occur concurrently in 20% of patients with SCI. Even more problematic than identifying the source of an infection is discerning whether fever is caused by an infection or a noninfectious condition that may closely mimic infection and cause almost one-fifth of episodes of fever in these patients.

A diagnostic conundrum may arise when unique thermoregulatory and autonomic disturbances cause fever in patients with SCI. Because of the imbalance between heat production and heat loss, patients with an injury to the spinal cord above T8 may not be able to maintain a normal body temperature in response to heating or cooling (poikilothermia). This phenomenon of altered thermoregulation is attributed to the loss of sweating and muscular activity below the spinal cord lesion. These factors may contribute to the occurrence of self-limited febrile episodes in patients with SCI that resolve spontaneously within hours to days. However, neither alterations in environmental temperature nor changes in a subject's sweating and muscular activity may explain the occurrence of prolonged fever in recently injured quadriplegic patients who have no identifiable focus of infection. This unique syndrome—so-called quadriplegia fever—lasts weeks to months and is problematic because it may incite repeated evaluation for infection and multiple courses of antibiotics, but to no avail. Rarely, fever may occur in the context of autonomic dysreflexia, a paroxysmal syndrome characterized mainly by hypertension, sweating, facial flushing, and headache. Occasionally, bradycardia may also be present and can help differentiate febrile episodes of autonomic dysreflexia from infection. This type of autonomic hyperactivity is seen only in patients with SCI above T6 and is usually triggered by distention of viscera (bladder and rectum), cutaneous stimulation (e.g., ingrown toenails), or even infection.

Treatment of infection in patients with SCI also poses special challenges. For example, two opposing factors resulting from changes in body composition after SCI can alter the disposition of certain systemically administered antibiotics such as vancomycin and aminoglycosides. On one hand, patients with SCI have expanded extracellular volume attributed to retention of extracellular water as subclinical edema and replacement of diminished skeletal muscle mass by extracellular water. As a result, these patients have a larger weight-adjusted volume of distribution of drugs and may require larger weight-adjusted loading and maintenance doses than able-bodied counterparts to achieve similar antibiotic concentrations. On the other hand, this potential effect on antibiotic concentration can be counteracted, at least in part, by the frequent overestimation of creatinine clearance when using current formulas that were originally devised for able-bodied individuals to predict creatinine clearance in patients with chronic SCI who have low serum creatinine levels. A proper 24-hour collection of urine can accurately estimate renal function. The multiple complicated conditions of patients with SCI result in frequent rehospitalization.

Urinary Tract Infection

UTIs are the most common type of infection in patients with both traumatic and nontraumatic SCI and occur at a rate of 2.5 episodes per patient per year. In general, inadequate antibiotics are available with increasing costs of treatment of UTI of about $2 billion a year in the United States. In patients with chronic indwelling bladder catheters (transurethral or suprapubic), bacteriuria is almost universal and occurs at a higher rate than in patients with intermittent bladder catheterization (98% vs. 70%). A longer interval between intermittent bladder catheterizations may be associated with a higher incidence of bacteriuria. Although outpatients may find it more practical to use clean reusable rather than sterile catheters for intermittent bladder catheterization, there is conflicting evidence regarding the value of clean versus sterile bladder catheterization.

Asymptomatic bladder colonization may progress to symptomatic UTI but often does not. Typical manifestations of UTI (including dysuria, urgency, frequency, suprapubic discomfort, and, in patients with pyelonephritis, costovertebral angle tenderness) are rarely present in patients with SCI. Instead, change in voiding habits, increase in the residual volume of urine in the bladder, foul-smelling urine, worsening of muscular spasticity, and aggravation of autonomic dysreflexia are often the only clinical clues to the presence of UTI. Because of the nonspecificity of these clinical manifestations, other causes should be excluded before diagnosing UTI. Although the lack of pyuria reasonably predicts the absence of UTI in patients with SCI, pyuria is a nonspecific finding that is also observed in uninfected individuals whose urinary tract is inflamed by catheter manipulations, renal calculi, and interstitial nephritis. Other commonly reported abnormal laboratory findings in the urine, including nitrite and leukocyte esterase, are also not specific for infection. Another diagnostic limitation is that about two-fifths of patients with SCI incorrectly attribute their bouts of illness to UTI. As with other patient populations who require bladder catheterization, bacterial concentration in urine of patients with SCI may not help differentiate between asymptomatic bladder colonization and symptomatic UTI. Most cases of UTI in patients with SCI are caused by commensal organisms of the bowel and perineum, particularly gram-negative bacilli and enterococci. The patient's sex and level of injury may affect the microbiology of organisms residing in the bladder. The finding of polymicrobial bacteriuria is particularly problematic in patients with SCI. Almost half of positive urine cultures in these patients grow more than one organism, and polymicrobial bacteriuria can be more prevalent in patients who have chronic indwelling urethral catheters. Although isolation of multiple bacterial species in the general population is often viewed as indicative of contamination, a similar finding in catheterized patients with SCI should not be disregarded. Isolation of several uropathogens can be associated with a UTI episode that fails to respond to antibiotic therapy directed against only one or some of the organisms but is eradicated after providing additional antimicrobial coverage against other isolated organisms. Although most urine cultures are obtained from patients with lower UTI only and in whom the yield of blood cultures is extremely low, the detection of concurrent bacteremia confirms the pathogenicity of organisms isolated from urine culture. Even in patients who have pyelonephritis in association with polymicrobial bacteriuria, isolation of only one organism from blood cultures may not negate the role of other bacteria in causing UTI. Because it is difficult to differentiate accurately between cultured organisms causing infection and asymptomatic colonizers, it may be reasonable to treat all potentially pathogenic organisms grown from urine cultures in patients with a diagnosis of infection.

A preventive approach that dramatically protects against UTI has not been identified since the advent of closed urinary drainage almost a half-century ago. Optimizing urinary drainage and switching, whenever feasible, from an indwelling bladder catheter to intermittent bladder catheterization or even external condom-based drainage, remain the cornerstone of prevention. The incidence of bladder stones, a condition associated with repeated bouts of UTI, is higher in patients who rely on indwelling versus intermittent bladder catheterization. Fig. 309.1 delineates the relationship between renal calculi, urinary tract obstruction, and infection. Upper tract urolithiasis is commonly managed with extracorporeal shock wave lithotripsy or percutaneous nephrolithotomy.

A number of approaches have been examined for their potential protection against infection. Although clinical data exist regarding the ability of antimicrobial-modified catheters (including hydrogel/silver-coated and nitrofurantoin-coated) and intermittent bladder catheters to reduce the rate of bacteriuria (not clinical UTI) in the general population, there is no strong evidence that such catheters can prevent or reduce clinical UTI. However, both a randomized controlled trial and a single-arm study reported that patients who received injection of botulinum toxin A into the detrusor muscle experienced about 45% reduction in the incidence of UTI compared with the preinjection period. Because botulinum toxin A reduces the detrusor pressure, it could help prevent reflux and therefore pyelonephritis. The coating of silicone with mannoside-poly(amido amine) decreases biofilm formation. Furthermore, a number of studies focused on the use of Escherichia coli strain 83972 to prevent UTI in patients with neurogenic bladder.

Asymptomatic bacteriuria can progress to symptomatic infection, and thus several approaches have been designed to prevent or eradicate asymptomatic bacteriuria. However, there were no significant differences in outcomes between high-dose and low-dose antibiotics given before endoscopic urologic procedures.