Published: October 2012
Infectious disease emergencies are conditions that have potential for significant harm to the patient if not recognized and treated promptly. Timely and appropriate intervention may significantly improve outcomes. The following is a discussion of important infectious disease emergencies.
Bacterial meningitis is an inflammation of the meninges caused by bacterial infection. Acute meningitis is characterized by the development of meningeal signs over the course of a few hours to a few days. The important causes of bacterial meningitis are outlined in Table 1.
A passive survey conducted in the United States between 1978 and 1981 revealed an annual incidence rate for bacterial meningitis of 3.0 cases per 100,000 population.1 During this period, bacterial meningitis was predominantly a disease of children, the most common offending pathogen being Haemophilus influenzae. The introduction of routine immunization of children against H. influenzae type B in the late 1980s dramatically reduced the incidence of infection with this microorganism. As a consequence, the overall incidence of bacterial meningitis–and particularly of meningitis caused by H. influenzae–decreased, so that bacterial meningitis is now a disease predominantly of adults.2
The initial event is usually nasopharyngeal colonization with a pathogenic microorganism.3 This is followed by mucosal invasion, bacteremia, and meningeal invasion.4 A marked inflammatory response occurs in the subarachnoid space, but this response is inadequate to control the infection. This inflammatory response results in increased permeability of the blood-brain barrier. This is responsible for the increased cerebrospinal fluid (CSF) protein content seen in patients with meningitis. Progression of meningitis leads to the development of cerebral edema, resulting in increased CSF pressure. Inflammation of blood vessels traversing the subarachnoid space may lead to their thrombosis. This can result in ischemia and infarction of the underlying brain.
*Mycobacterium tuberculosis is also a bacterium and can cause meningitis, but it is usually discussed separately as tuberculous meningitis.
Patients with acute bacterial meningitis usually present with headache, neck stiffness, fever, projectile vomiting, and photophobia. In more advanced disease, there is progressive clouding of consciousness. On examination, neck rigidity may be seen and Kernig’s and Brudzinski’s signs may be elicited. Cranial nerve palsies or focal neurologic signs may be seen in a minority of patients. The presence of petechial skin lesions should raise suspicion for meningococcemia.
The differential diagnosis includes viral and tuberculous meningitis, viral meningoencephalitis, subarachnoid hemorrhage, and primary amebic meningoencephalitis. Differentiation from viral meningitis on clinical grounds is usually difficult, and requires laboratory testing. Where tuberculosis is prevalent, it must be recognized that tuberculous meningitis can sometimes manifest acutely and could be mistaken for bacterial meningitis. Viral meningoencephalitis may manifest somewhat similarly with headache and fever, but patients would usually have more profound alteration in the sensorium early in the illness and neck stiffness may not be prominent. The most prominent symptom of subarachnoid hemorrhage is severe headache with a rapid onset. Primary amebic meningoencephalitis is a rare condition with a presentation similar to that of acute bacterial meningitis, but cultures are negative and amebae can be detected in the CSF by careful microscopic examination. There is usually a recent history of swimming in a warm freshwater lake or pond.
The most important diagnostic test is a lumbar puncture, which should always be performed in all patients with suspected acute meningitis. Imaging tests do not help in making the diagnosis or identifying the cause of bacterial meningitis. It is not necessary to obtain a computed tomography (CT) scan before performing a lumbar puncture unless there are focal neurologic deficits.5 The CSF should be sent for cell count, protein and glucose levels, and Gram staining and culture. Typical CSF findings in acute bacterial meningitis include an elevated opening pressure, increased CSF white blood cell (WBC) count (100-10,000 cells/mcL), usually with a predominance of neutrophils, increased CSF protein level (> 50 mg/dL), and decreased CSF glucose level (< 40% of simultaneously measured serum glucose level).6 Gram staining may reveal the presence of microorganisms and, if so detected, would be helpful for guiding therapy. In viral meningitis, the CSF WBC count is elevated, but the cells are usually predominantly lymphocytes, and the CSF glucose level may be normal or marginally decreased. The best way to confirm a diagnosis of viral meningitis is by specific polymerase chain reaction (PCR) testing, if available.
The management of bacterial meningitis includes appropriate antibiotic therapy and adjunctive corticosteroids.7,8 Ideally, the lumbar puncture should be done before the administration of antibiotics. However, antibiotic administration should not be delayed for any reason if the lumbar puncture cannot be immediately performed. A lumbar puncture should be performed as soon as possible, even if antibiotics have already been administered; the possibility of being able to make a definite causal diagnosis, and its value in guiding subsequent therapy and managing possible complications, are fully worth the effort. Empiric antibiotics should be selected based on the expected pathogens. The patient’s age, presence or absence of risk factors such as middle ear or sinus disease, or recent neurosurgery provide clues about the cause and pathogenesis.
It is recommended that patients be started on adjunctive dexamethasone 10 mg IV every 6 hours for 4 days with the first dose of antibiotics. This has been shown to improve outcomes in patients with bacterial meningitis.7 Antibiotic selection and dosing should also take into consideration the agent’s ability to cross the blood-brain barrier and achieve an effective concentration in the CSF. In adults, initial empiric treatment should provide adequate therapy for Streptococcus pneumoniae and Neisseria meningitidis. Increasing resistance of S. pneumoniae to beta-lactam antibiotics (including ceftriaxone) has prompted recommendations to initiate empiric antibiotic therapy with a regimen consisting of vancomycin and ceftriaxone.6 If Listeria monocytogenes is a possibility (eg, in older adults, pregnant women, and those with cellular immune deficits), ampicillin should be added. If Pseudomonas aeruginosa is a possibility, as after neurosurgical procedures, ceftazidime should be used instead of ceftriaxone. Antibiotic therapy should be adjusted once the causative microorganism has been identified. Duration of therapy for bacterial meningitis has not been adequately defined. For meningococcal meningitis, 7 days of therapy is considered adequate. S. pneumoniae should be treated for 10 to 14 days. L. monocytogenes should be treated for at least 21 days.9
Acute meningococcemia is a disseminated infection caused by Neisseria meningitidis, with high mortality rates in those with fulminant disease. Meningococcal infection occurs in an endemic pattern, with periodic epidemics. There are substantial cyclic variations in disease incidence. In the United States, epidemics account for less than 5% of the reported cases. The incidence of meningococcal disease in the United States peaked at 1.7 cases per 100,000 population in 1997.10
The pathogenesis of meningococcal infection begins with nasopharyngeal colonization. About 10% of the population has asymptomatic nasopharyngeal carriage of N. meningitidis during nonepidemic periods. A small proportion of carriers go on to develop invasive meningococcal disease. People who develop invasive disease generally do so soon after acquisition of carriage.11 Factors that facilitate invasive disease include agent factors, such as virulence, and transmissibility and host factors.
Patients with deficiencies of the late components of the complement pathway are at markedly increased risk of developing recurrent episodes of meningococcal infections.12 Genetic variants of mannose-binding lectin (MBL), a plasma opsonin that initiates the MBL pathway of complement activation, may also make patients more susceptible to meningococcal infections.13
Meningococcal infection can manifest in a variety of forms: bacteremia without sepsis, meningitis (with or without meningococcemia), acute meningococcemia (with or without meningitis), a meningoencephalitic picture, or chronic meningococcemia. The most fulminant form is acute meningococcemia, in which death may ensue within hours of the onset of symptoms.
The most common manifestation of acute meningococcemia is fever with rash. The rash usually begins with petechiae, initially with a few discrete lesions 1 to 2 mm in diameter that often progress and coalesce to form larger ecchymotic lesions (Figure 1). In cases of associated meningitis, meningeal signs and symptoms may also be present.
The shock state is a dominant feature in patients with acute meningococcemia, and is often accompanied by disseminated intravascular coagulation (DIC). Meningococcemia can lead to complications such as massive adrenal hemorrhage; DIC; arthritis; heart problems such as pericarditis and myocarditis; neurologic problems such as deafness and peripheral neuropathy; and peripheral gangrene.14 In epidemic settings in Third-World countries, case-fatality rates as high as 70% have been recorded. In endemic settings in industrialized countries, the mortality rate is approximately 8%, but might be as high as 19%.
Meningococcemia does not always manifest in a fulminant manner. An unusual manifestation of meningococcal infection is chronic meningococcemia, which manifests with low-grade fever, rash, and arthritis. This manifestation is identical to that of chronic gonococcemia.
When patients present with an acute febrile illness with the characteristic ecchymotic rash, the diagnosis is not difficult to make. Early infection could be missed if a careful physical examination is not carried out in a patient with an acute febrile illness. Definitive diagnosis requires isolation of the microorganism from a normally sterile site. Samples for blood cultures should be obtained before the administration of antibiotics, if possible. Antibiotic therapy rapidly sterilizes the blood and CSF in patients with meningococcal infection.15,16 CSF cultures are often positive for microorganisms, even in patients who do not have clinical evidence of meningitis,17 and should always be examined when meningococcemia is suspected. Microorganisms may also be identified in the biopsy of petechial skin lesions.
The treatment of acute meningococcemia involves appropriate antibiotic therapy, along with supportive therapy for shock, heart failure, DIC, and other complications. Early antibiotic therapy has been conclusively shown to improve outcomes in patients with meningococcal disease.18 The recommended treatment for severe meningococcal infection is a third-generation cephalosporin with good CSF penetration. Ceftriaxone, 1 g every 12 hours, is the most commonly used treatment.19 Cefotaxime and ceftazidime should be equally efficacious alternatives. Patients who are allergic to cephalosporins may be treated with chloramphenicol, 100 mg/kg, in 4 divided doses. The maximum total dosage is 4 g/day.20 High doses of penicillin G should also usually be adequate; however, small numbers of resistant N. meningitidis have been reported, and therefore penicillin G is not ordinarily the first-choice antibiotic in the absence of susceptibility data. The shock state is a dominant part of the clinical picture of meningococcemia and supportive management is important. The use of steroids for meningococcemia is controversial, and a recommendation for the routine use of steroids for treatment of this condition cannot be made.
Household contacts are at significantly higher risk of infection.21 Chemoprophylaxis is recommended for household contacts, daycare center staff and clients, and anyone exposed to the patient’s oral secretions. Among health care workers, this includes people who intubated the patient and who provided suction to clear secretions. Effective prophylactic treatments include a single 1-g dose of ceftriaxone intravenously or intramuscularly, a single 500-mg dose of ciprofloxacin, a single 500-mg dose of azithromycin, and 600 mg of rifampin every 12 hours for 2 days.18
Subdural empyema is a condition in which there is collection of pus in the region between the dura and the arachnoid. The most common causes of subdural empyema are aerobic and anaerobic streptococci (especially the S. milleri group), Staphylococcus aureus and, to a lesser extent, aerobic gram-negative bacilli.22,23 Studies have found anaerobic infections in varying proportions of infections. However a high proportion of patients have had anaerobic microorganisms recovered.24 This raises the possibility that these infections are usually polymicrobial, with anaerobic microorganisms usually present. Subdural empyemas account for 15% to 20% of all localized intracranial infections.23
Cranial subdural empyema is usually a complication of infection of the paranasal sinuses.23 Less commonly, it results from spread from the middle ear.25 It may also occur as a complication of trauma or neurosurgery. Infection spreads intracranially through the emissary veins that communicate between the veins draining the facial structures and intracranial venous channels. In a small proportion of cases, subdural empyema may occur by metastatic spread, usually from a pulmonary infection, for an unexplained reason.
Patients with this condition usually present acutely, with headache and vomiting. Most patients have an altered mental status at presentation and the level of consciousness deteriorates rapidly. Patients may have neurologic deficits and complications such as local cerebritis, cerebral abscesses, and septic dural venous thromboses may occur.
The diagnosis should be considered in any patient who presents with features suggestive of meningitis and a focal neurologic deficit or rapid deterioration in the level of consciousness. If recognized, lumbar puncture should not be performed because of the risk of cerebral herniation.23 CSF findings would be nonspecific, with an elevated opening pressure, neutrophilic or mixed pleocytosis, and elevated protein level. Gram staining and culture of CSF are usually negative. The diagnostic procedure of choice is magnetic resonance imaging (MRI). If not available, CT scanning with contrast should be done. CT is inferior to MRI in detecting empyemas at the base of the brain and in the posterior fossa.
Effective treatment for cranial subdural empyema requires a combined surgical and medical approach. Empiric antibiotic therapy should be broad spectrum and include coverage for gram-positive pyogenic bacteria and anaerobes. Vancomycin is a reasonable choice. Cultures obtained at the time of surgery will help tailor antibiotic treatment. The goal of surgery is complete evacuation of the purulent collection, which may be accomplished by craniotomy or through burr holes, depending on the circumstances of the case. It is important to evacuate the collection completely, and a craniotomy or multiple surgical procedures may be necessary to accomplish this. Up to 50% of patients who are treated with burr hole drainage require reoperation, compared with 20% of those treated with craniotomy.26 The duration of antibiotic therapy is usually 3 to 4 weeks after adequate drainage. If there is associated osteomyelitis of the skull, treatment should be extended to approximately 6 weeks.
This term encompasses several specific clinical entities characterized by disease processes that produce necrosis of subcutaneous tissue, muscle, or both, and that progress rapidly and require a combined emergent surgical and medical approach for optimal outcomes. These entities include necrotizing fasciitis, streptococcal necrotizing myositis, clostridial myonecrosis (gas gangrene), and nonclostridial crepitant myositis.
Type I necrotizing fasciitis is a mixed infection caused by an anaerobic bacterium (usually Bacteroides or Clostridium) in association with a facultative anaerobic microorganism, such as a streptococcus or a member of the Enterobacteriaceae. Type II necrotizing fasciitis, hemolytic streptococcal gangrene, is caused by group A, B, C, or G streptococci. Other microorganisms may be present in the mix. Community-associated methicillin-resistant S. aureus (CA-MRSA) has recently been described as a cause of necrotizing fasciitis.27
Clostridial myonecrosis, also commonly known as gas gangrene, and streptococcal necrotizing myositis, as their names imply, are caused by Clostridium spp. and by beta hemolytic streptococci, respectively.
Nonclostridial crepitant myositis encompasses several clinical entities that may result from mixed infection caused by: anaerobic streptococci, along with group A streptococci or S. aureus (anaerobic streptococcal myonecrosis); a mixture of anaerobic and facultatively anaerobic microorganisms (synergistic nonclostridial anaerobic myonecrosis, or Meleney’s bacterial synergistic gangrene); Proteus spp., Bacteroides spp., and anaerobic streptococci in devitalized limbs (infected vascular gangrene); Vibrio vulnificus; and Aeromonas hydrophila.
As a group, these illnesses are uncommon but not rare, and prompt recognition and appropriate management can significantly improve outcomes.
Necrotizing fasciitis usually begins with the introduction of the offending microorganism into the subcutaneous structures, usually as a result of minor trauma. Gas gangrene occurs in situations in which muscle injury is compounded by wound contamination with soil or other foreign material harboring spores of a tissue-invasive Clostridium, such as C. perfringens, C. novyi, and C. septicum. Such injuries include war injuries, compound fractures, and septic abortion. Most cases of streptococcal myositis appear to begin spontaneously. The different forms of nonclostridial myonecrosis usually begin with the introduction of the offending microorganisms at the time of usually minor trauma. Aeromonas hydrophila myonecrosis occurs as a result of inoculation of the microorganism at the time of penetrating injury in a freshwater setting or in association with fish or other aquatic animals. In all these conditions, there is rapid progression of disease, often with gas formation in the muscles and subcutaneous tissues, and in many cases associated with the development of gangrene.
Diabetes mellitus is the most important risk factor for the development of necrotizing soft tissue infections.28 Other risk factors include alcoholism, corticosteroid use, and parenteral drug use.
Necrotizing fasciitis is usually an acute process, with severe infection of the superficial and deep fascia. It most commonly occurs in the extremities. The affected area becomes erythematous, swollen, warm, and painful. The infection typically progresses rapidly, with the skin becoming darker, and over a few days bullae and skin breakdown develop. In the polymicrobial form, crepitations may be felt subcutaneously, indicating the presence of gas. Development of anesthesia over an erythematous area may precede development of skin breakdown and may serve as a warning sign that the disease process is more serious than cellulitis. Pain out of proportion to the skin changes also may be an indicator of a more serious infection. On palpation, the affected area has a woody hard feel. Increasing tissue edema may lead to the development of compartment syndrome.
Necrotizing myositis or myonecrosis may occur without overt findings on the skin surface. The predominant symptom is intense muscle pain, usually accompanied by fever. Patients usually appear more ill than would be expected from the physical findings. Gas gangrene and other syndromes of necrotizing myositis caused by anaerobic microorganisms will also have crepitations because of the presence of subcutaneous gas.
At initial presentation, it may not be possible to make a clinical distinction between necrotizing fasciitis and necrotizing myositis. Indeed, both processes may occur simultaneously, especially with streptococcal infection. Lack of involvement of the overlying skin does not exclude the presence of an underlying necrotizing process.
Streptococcal necrotizing soft tissue infections are usually associated with the toxic shock syndrome. Acute vascular compromise from trauma or embolic occlusion leads to tissue infarction and may progress to infected vascular gangrene if the appropriate microorganisms gain access to the devitalized tissues.
Clinical suspicion is important in order to make an early diagnosis of a necrotizing soft tissue infection. A clue to the presence of a deep necrotizing process is the presence of tenderness clearly beyond the areas of apparent involvement in the skin. Leukocytosis is common. The creatine kinase (CK) level is usually elevated, but may be normal in cases of necrotizing fasciitis with minimal muscle involvement. Ultrasonography, CT scanning, or MRI will usually reveal muscle swelling and fluid in muscle compartments, but may not be apparent early in the disease process. Histopathologic examination will reveal the presence of sheets of neutrophils in fascial planes. Gram staining of tissue exudates will reveal the presence of microorganisms.
It is not always obvious whether a skin or soft tissue infection is a necrotizing one. When considered a possibility, aggressive management is important. Clinical features cannot accurately predict the causative microorganism. A prudent approach would be to treat with antibiotics that are effective against group A streptococci, S. aureus, enteric gram-negative bacteria, and anaerobic microorganisms until the etiologic diagnosis has been established. The antibiotics of choice for initial empiric therapy are clindamycin plus ampicillin-sulbactam plus ciprofloxacin.29 If there is reason to suspect MRSA infection, vancomycin should be added. Antibiotic therapy should be modified when culture and susceptibility data become available. A lack of response to a reasonable trial of antibiotics should prompt emergent surgical intervention. Prompt and aggressive fasciotomy and debridement of devitalized tissue are necessary to gain control of the infection. Early surgical intervention reduces mortality.28 If infection is advanced, amputation may be necessary and lifesaving.
Toxic shock syndrome (TSS) is a severe toxin-mediated bacterial disease characterized by shock resulting from an excess of inflammatory cytokines. Two important syndromes, staphylococcal TSS and streptococcal TSS, are recognized to be caused by S. aureus and Streptococcus pyogenes, respectively. Both are uncommon diseases. The incidence rate of streptococcal TSS in the United States is 3.5 per 100,000 population per year. Staphylococcal TSS has an overall incidence of about 1 per 100,000, with menstrual TSS about twice as common as nonmenstrual TSS.30 At the peak of the epidemic of menstrual TSS, before the recognition of the association between the use of certain tampons and TSS, the incidence rate of menstrual TSS was as high as 10 per 100,000 population per year and accounted for over 90% of all cases of staphylococcal TSS.
Toxic shock syndrome is a toxin-mediated disease.31 Several exotoxins of S. aureus and S. pyogenes are capable of stimulating excessive T cell responses, and are thus known as superantigens. These toxins include toxic shock syndrome toxin-1 (TSST-1) and staphylococcal exotoxins A, B, and C (SEA, SEB, SEC) of S. aureus, and streptococcal pyrogenic exotoxins A, B, and C (SPEA, SPEB, SPEC) of S. pyogenes. These toxins are capable of binding both major histocompatibility complex (MHC) class II molecules of antigen-presenting cells and the Vβ region of T cell receptors, leading to broad-range induction of T cell proliferation. The resulting excessive production of inflammatory cytokines (interleukin-1 and -6 [IL-1, IL-6], tumor necrosis factors α and β [TNF-α, TNF-β], interferon gamma [IFN-γ]) leads to increased capillary permeability resulting in tissue damage to various organs and shock.
Staphylococcal TSS is commonly associated with menstruation (menstrual TSS).32 The pathophysiology of menstrual TSS includes a high local protein level and relatively high local pH (caused by the presence of blood and blood products), high partial pressure of carbon dioxide (caused by higher than atmospheric Pco2 in blood), and high Po2 (introduced by high-absorbency tampons).31 The production of TSST-1 by colonizing S. aureus is stimulated in such an environment. Nonmenstrual TSS can be caused by S. aureus infection at any site in the body, including surgical wounds, the lungs, peritoneal dialysis catheters, and skin and mucosal infections. The illness is mediated by TSST-1 or SEA or SEB produced by the microorganisms at the site of infection.
Toxic shock syndrome manifests as a multisystem illness, with shock being a prominent feature.31 Clinical features include high fever, hypotension, tachycardia, tachypnea, anasarca, and a morbilliform rash. Many patients also have myalgias and gastrointestinal symptoms, such as vomiting, abdominal pain, and diarrhea. Patients may develop confusion. The disease progresses rapidly and, especially with streptococcal TSS, can lead to death within 24 to 48 hours. Menstrual TSS starts within 2 days of the beginning or end of menses in women using high-absorbency tampons.32 In many patients with nonmenstrual TSS, the site of infection may show minimal inflammation and may not be readily apparent.
Streptococcal TSS is generally a more serious condition.33 In this condition, the site of infection with S. pyogenes may be obvious. It is usually a necrotizing soft tissue infection, but streptococcal TSS has been described in patients with pneumonia, meningitis, septic arthritis, peritonitis, and other deep infections.33 Patients are usually very ill and may develop acute respiratory distress syndrome (ARDS) or DIC. The mortality of adequately treated staphylococcal TSS is about 5%. The mortality of streptococcal TSS is about 50%.
The diagnostic criteria for TSS are outlined in Tables 2 and 3.34,35 In streptococcal TSS, streptococci can usually be detected in culture at the affected site or in blood culture. In staphylococcal TSS, it is rare to detect staphylococci, except if vaginal cultures are obtained from patients with menstruation-associated TSS. Prompt diagnosis requires recognition of the constellation of symptoms and signs, with confirmation by additional laboratory testing to look for abnormalities that indicate damage to the organ systems expected to be involved by the process. CT or MRI helps in defining the presence of deep soft tissue infection in patients with streptococcal TSS. Gas is not produced, but the diagnosis should not be excluded if imaging findings do not appear impressive when clinical features are suggestive of the disease.
Diagnostic criteria include:
Plus multisystem involvement (≥ 3 of the following):
Negative results on the following tests, if performed:
Desquamation 1-2 wk after onset of illness, particularly over palms and soles
*All 6 criteria have to be satisfied to make a definite diagnosis. Fulfillment of the first 5 criteria makes a probable diagnosis.
Diagnostic criteria include 1 of the following:
Plus 2 or more of the following:
*All 3 criteria must be met. If S. pyogenes is isolated from a normally sterile site, the diagnosis is definite; if isolated from a nonsterile site, the diagnosis is probable.
Management of TSS includes eradication of the focus of infection plus supportive care, which includes fluid resuscitation and vasopressors, as necessary.36 Large volumes of crystalloids may be required because of the loss of intravascular volume caused by capillary leak. Circulating bacterial hemolysins may lead to moderate to severe anemia, necessitating blood transfusions. When the focus of infection is identified in staphylococcal TSS, it is important to drain abscesses and treat with appropriate antibiotics. If a hyperabsorbent tampon is in place, it should be removed. Patients who have streptococcal TSS do better with combinations of clindamycin and cell-wall active agents compared to patients who receive cell-wall active agents alone.37 Because clindamycin is not affected by bacterial inoculum or stage of growth, and because it inhibits synthesis of bacterial toxin, there are theoretical reasons why it would also be effective in these patients. Thus, in both staphylococcal and streptococcal toxic shock syndrome, clindamycin should be included initially.
Staphylococcal TSS should also initially be treated with vancomycin. Antibiotic therapy can be modified once susceptibility data become available. MRSA infections should be treated with vancomycin; MSSA infections with oxacillin; and penicillin-susceptible S. aureus should be treated with penicillin G. Treatment of streptococcal TSS usually includes aggressive surgical debridement, with antibiotic therapy and supportive care. The antibiotic of choice is penicillin G.
Fever in a neutropenic patient is defined as a single oral temperature > 101°F (38.3°C) or a temperature >100.4° (38.0°C) sustained over a 1-hour period.38 Neutropenia is defined as an absolute neutrophil count (ANC) < 500 cells/mcL and/or ANC that is expected to decrease to <500 cells/mcL over the next 48 hours.38 More than 50% of neutropenic patients who develop a fever have an obvious or occult infection. The risk of invasive infection increases with the degree of neutropenia. At least 20% of patients with an absolute neutrophil count < 100 cells/mcL will develop bacteremia.
Neutrophils and macrophages represent the cellular arm of innate immunity. Neutrophils are recruited to the site of infection, where their role is to ingest the offending microorganisms. Patients with neutropenia have decreased neutrophil numbers, and their inflammatory responses are blunted. These patients may therefore have skin and skin structure infections, with minimal erythema or induration, urinary tract infection without pyuria, and pulmonary infection without chest radiographic infiltrates. The most likely sources of bacterial invasion in such patients are the skin, because of the breaching of skin integrity by invasive vascular catheters, and the gastrointestinal tract, because of mucosal damage from chemotherapeutic agents. The most common bacterial pathogens causing fever in patients with neutropenia are listed in Table 4.
Viridans group streptococci
Adapted from Freifeld AG, Bow EJ, Sepkowitz KA, et al: Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis 2011;52:e56-e93. Copyright ©2011 Infectious Disease Society of America, with permission from Oxford University Press.
Patients may present with fever only. The clinical presentation may appear to be relatively mild, especially to the inexperienced observer. Furthermore, many patients with chemotherapy-induced neutropenia may be taking corticosteroids, which may actually mask fever. In neutropenic patients who are taking corticosteroids, hypothermia, hypotension, or other unexplained acute clinical deterioration should be considered as a fever equivalent.
A careful physical examination should be carried out to identify any possible sources of infection. The clinician should obtain 2 or 3 sets of blood cultures, urinalysis and urine culture and, if the patient has productive cough, sputum Gram staining and culture should be performed. Chest radiographs should be obtained and there should be a low threshold for obtaining a CT scan of the chest because the latter is significantly more sensitive for detecting pulmonary processes. Other investigations should be carried out if there are any localizing symptoms or signs. Meningitis is uncommon and lumbar puncture should only be considered if there is strong evidence of meningitis.
For patients with neutropenic fever, empiric therapy should provide adequate coverage against enteric gram-negative bacteria and P. aeruginosa. If S. aureus infection is considered a possibility, as in patients with indwelling vascular catheters, vancomycin should also be given until culture data become available. For patients with catheter-related bloodstream infections it is usually necessary to remove the catheter.39
Several antibiotic regimens are effective for the empiric treatment of neutropenic fever, but no specific regimen is recommended for all patients. Criteria have been defined to identify those patients considered to be at high-risk for progression to severe illness.40 Patients deemed to be low-risk may be treated with oral ciprofloxacin plus oral amoxicillin-clavulanate. Those deemed to be high–risk–all patients with an ANC of <100 cells/mcL, duration of neutropenia exceeding 7 days, peak temperature 102.2° F (39° C) or higher, active malignancy, lack of early evidence of marrow recovery, and any obvious focus of infection–should be hospitalized and treated with IV antibiotic therapy consisting of monotherapy with an anti-pseudomonal Β-lactam agent such as cefepime or ceftazidime, a carbapenem (meropenem or imipenem), or piperacillin-tazobactam.38 Other antimicrobials may be added based on clinical circumstances. Recent infection or colonization, or treatment in a hospital with high rates of endemicity, may necessitate addition of antimicrobial agents that would specifically treat MRSA, vancomycin-resistant Enterococcus (VRE), extended-spectrum Β-lactamase (ESBL)-producing gram-negative bacteria, and carbapenem-resistant Klebsiella pneumoniae (CRKP).38
Antibiotic therapy should be modified if cultures identify a causative microorganism. If a gram-positive infection is not identified by day 3, an empirically started gram-positive bacterial agent such as vancomycin or daptomycin may be stopped. If there is worsening after 3 days, a change in the antibiotic regimen should be considered. If the fever persists after 4-7 days, empiric antifungal therapy should be initiated.
The duration of therapy must be based on the clinical response and clinical status of the patient. It is clear that neutropenic patients are a heterogeneous group. Patients who have undergone allogeneic stem cell transplantation, who have had graft-versus-host disease, severe mucosal erosive disease, extensive prior antibiotic treatment, or prolonged hospitalization are at higher risk of subsequent infectious complications.41 Patients who are neutropenic for fewer than 7 to 10 days are at low risk of infectious complications. If the fever resolves in 3 to 5 days, antibiotics may be stopped 2 days after the resolution of fever if the ANC is ≥ 500 cells/mcL for at least 48 hours. Antibiotic therapy can be stopped at 5 to 7 days after the resolution of fever in low-risk patients if the ANC is ≤ 500 cells/mcL, but should be continued for at least 2 weeks in high-risk patients with an ANC ≤ 500 cells/mcL. If the fever persists beyond 5 days, but there is no obvious identified infection, antibiotics may be stopped 4 to 5 days after the ANC recovers above 500 cells/mcL, but should be continued for at least 2 weeks if the ANC remains below 500 cells/mcL.
The importance of sepsis in these patients is the potential for progression from a healthy state to death within 24 hours when infected with certain microorganisms. This situation is not common but, when it does occur, the mortality rate is high. It is much more likely to occur at younger ages, with infection rates of 15.7% in infants, 10.4% in children younger than 5 years, 4.4% in children younger than 16, and 0.9% in adults.42 The lifetime risk of sepsis following splenectomy has been estimated to be approximately 5%.43 The risk of infection persists throughout life and is highest in the first 2 years after splenectomy.42
The microorganisms that cause deadly infections in patients with prior splenectomy or functional hyposplenism are S. pneumoniae, H. influenzae, N. meningitidis, Capnocytophaga canimorsus, and Babesia microti or B. divergens. The most common responsible microorganism is S. pneumoniae, accounting for 50% to 90% of cases.42,44
The innate immune system provides immediate protection against invasion by microorganisms. The adaptive immune system provides a more vigorous, targeted, and durable immune response by producing specific antibodies. However, the antigenic priming process for this response takes time, and makes this mechanism protective in subsequent infections, not in initial infections. B–1–a B cells are a subset of B lymphocytes that produce antibodies without prior specific exposure.45,46 These are natural antibodies of the immunoglobulin M (IgM) isotype that target polysaccharides of bacteria.47 Thus, they are part of the innate immune system and are responsible for the early inhibition of encapsulated bacteria. Splenectomy results in a marked reduction in the number of B-1a B cells.48 This reduction seriously compromises the ability of the body to mount an early response to infection with encapsulated microorganisms such as S. pneumoniae, H. influenzae, and N. meningitidis.
The spleen is also responsible for the clearance of erythrocytes with abnormal inclusions from the circulation. Such abnormal inclusions include intra-erythrocytic parasites such as Babesia species. Most cases of infection with Babesia have occurred in splenectomized individuals.49 The removal of erythrocytes infected with malarial parasites is also affected by splenectomy, but whether this increases the severity of disease is unclear.
The initial presentation, if patients present early, may be no different from that of a febrile illness in an otherwise healthy patient. Typical symptoms are fever, chills, sore throat, vomiting or diarrhea, and diffuse muscle aches. In young children, meningitis is common. Rapid progression within hours is the usual course. Shock, seizures, and DIC often accompany the clinical deterioration. Purpura fulminans and extremity gangrene can occur. With modern treatment, the mortality rate for sepsis following splenectomy is in the range of 50% to 70%.42,44
In the setting of prior splenectomy or functional hyposplenism, any febrile illness must raise suspicion for sepsis. The clinician should obtain 2 to 3 sets of blood samples for culture. Other investigations should be performed if there are any localizing symptoms or signs. In children, the CSF should be examined. The degree of bacteremia is usually several orders of magnitude greater than in patients with a functional spleen, and blood cultures usually turn positive within 12 to 24 hours. Because of the degree of bacteremia, Gram staining of the buffy coat often reveals the presence of bacteria. Bacteria may even be seen on a peripheral blood smear, which will also show the presence of Babesia. Thus, a peripheral smear should always be examined in any patient with postsplenectomy sepsis.
The most important aspect of management of sepsis following splenectomy is to recognize the condition and the potential for its occurrence in any patient who has had a splenectomy and who presents with a febrile illness. Antibiotics should be started immediately. Antibiotic therapy should not be withheld for completion of blood cultures. Empiric antibiotic therapy should consist of vancomycin and ceftriaxone. This empiric regimen will adequately treat all the pathogens definitely associated with sepsis in these patients, except Babesia. Vancomycin may be stopped if resistant pneumococci are not isolated. In those rare cases in which Babesia is the causative agent, treatment is with quinine plus clindamycin.
All patients should receive the 23-covalent bond, unconjugated, capsular pneumococcal polysaccharide vaccine (PPV23) at least 2 weeks before elective splenectomy.50 If the splenectomy is an emergent procedure, the vaccine should still be given as soon as possible. Revaccination once after 5 years is also recommended.51 Although there are fewer data on the protective efficacy of the H. influenzae type B (Hib) vaccine and meningococcal vaccine, both should be administered as soon as possible, preferably before splenectomy.
Children are often given prophylactic penicillin V for the first few years after splenectomy, which may provide some protection. This approach is not recommended for adults because of the uncertainty of the magnitude of protection offered and the potential for selection of resistant microorganisms.
Malaria in humans is caused by 5 species of Plasmodium: P. falciparum, P. vivax, P. malariae, P. knowlesi, and P. ovale. Malaria should always be considered in any person presenting with fever who has recently been in a malaria-endemic region, even if malaria prophylaxis was taken. Malaria occurs in most of the tropical and subtropical regions of South and Southeast Asia, sub-Saharan Africa, and South America. P. falciparum malaria is far more dangerous than the non–P. falciparum malarias. In 2009, there were an estimated 225 million cases of malaria worldwide.52 Globally, more than 1 million deaths occur from malaria each year.53 The vast majority are caused by P. falciparum. P. knowlesi is a newly recognized human malarial parasite that is capable of causing severe malaria.54 The following discussion focuses on P. falciparum malaria, the most important cause of malaria worldwide.
The life cycle of Plasmodium involves an asexual stage in a mosquito belonging to the genus Anopheles and its sexual stage in humans. Humans are infected by the bite of an infected mosquito. In humans, the infective sporozoites initially invade the liver, develop into merozoites in the liver cells, and are released into the bloodstream after several days. P. vivax and P. ovale, in addition to undergoing immediate development in the liver, are also capable of remaining in liver cells as latent infections for several months to years; the latent forms are known as hypnozoites. These hypnozoites perpetuate what is known as the exoerythrocytic cycle, its importance being that effective treatment for these infections must include medications active against the liver stage of the parasite.
This cycle is not seen with P. falciparum. The merozoites released from the liver cells invade red blood cells, where they develop into trophozoites that mature into schizonts, which then divide into several merozoites. The merozoites are in turn released by lysis of the infected cells; the process takes 48 hours in the case of P. falciparum. Lysis of infected cells occurs in a coordinated fashion every 24 hours, with 2 batches undergoing lysis on alternate days. It is at this time that the patient experiences chills and rigors.
The released merozoites are capable of infecting other red blood cells, thus perpetuating illness. Some schizonts mature into male or female gametocytes. For transmission of malaria to occur, gametocytes must be ingested by an appropriate mosquito. Fertilization occurs in the lumen of the mosquito stomach; the resulting zygote penetrates the wall of the stomach and develops into an ookinete on the outer surface of the stomach wall. When the ookinete matures, it develops into an oocyst that multiplies and ruptures to release many sporozoites into the body cavity of the mosquito. The sporozoites then find their way to the mosquito’s salivary glands, where they are poised to be transmitted to another human host at the time of the mosquito’s next blood meal.
The most important aspect of the pathogenesis of disease caused by P. falciparum is the ability of the parasite to sequester in the deep microvasculature. P. falciparum erythrocyte membrane protein-1 (PfEMP-1) is a molecule expressed on the surface of infected erythrocytes. This molecule is capable of adhering to various host cell surface ligands, including CD36 on endothelium, monocytes, and platelets and intercellular adhesion molecule-1 (ICAM-1) on endothelial surfaces.55 Furthermore, PfEMP-1 also mediates rosetting of host erythrocytes.56 This combination of endothelial binding and rosette formation leads to massive sequestration of parasitized erythrocytes in the cerebral, renal, hepatic, and other microcirculations, leading to the complications of P. falciparum malaria.
The most important symptom of malaria is fever. Classic descriptions of malaria include the occurrence of paroxysms that begin with chills and rigors, followed by high fever and then by profuse sweating and defervescence, with the entire paroxysm lasting a few hours. These symptoms classically occurred every 24 hours in patients with P. falciparum malaria. In our current perspective, paroxysms may not be very prominent, and patients may even have continuous fever. Headache is common, and may be seen even in the absence of cerebral malaria.
P. falciparum malaria is far more likely to lead to complications than is the non–P. falciparum malarias. The most important complications of P. falciparum malaria are severe anemia, hypoglycemia, cerebral malaria, acute renal failure, ARDS, acute hepatocellular failure, and DIC.
Several clinical features indicate specific end-organ damage in P. falciparum malaria; these include headache, seizures, and loss of consciousness (with cerebral malaria), decreased urine output or blood in urine (blackwater fever), deep jaundice (algid malaria), bleeding from multiple sites (caused by DIC), and respiratory failure (caused by ARDS).
Malaria must be considered in the differential diagnosis of any febrile illness in a patient living in or having recently traveled to an area in which malaria is endemic. The most important diagnostic and most practical test is the Wright’s- (or Wright’s-Giemsa) stained peripheral blood smear (thick and thin), which should be examined immediately if malaria is suspected. These tests can be performed in any laboratory but the limitation in nonendemic areas may be the nonavailability of personnel with experience in reading the slides. The most likely finding in patients with P. falciparum malaria are ring-shaped trophozoites within red blood cells (Figure 2). Antigen-based tests and serologic tests also exist for the diagnosis of malaria. While these tests do not require much expertise to read, they are unlikely to be available in areas in which malaria is not endemic.
All patients with P. falciparum malaria should be hospitalized, at least for the first 48 hours after initiation of treatment, regardless of how well they may appear at presentation. Patients with severe malaria may require ICU admission. The effectiveness of different antimalarials depends on the geographic area in which the infection was acquired. Treatment regimens effective against P. falciparum are outlined in Table 5. The most commonly used drugs are quinine or artemisinin derivatives, which should be effective for almost all cases of malaria in most regions. In many endemic regions, there is significant resistance to pyrimethamine-sulfadoxine, and this drug should be used with caution; if used, patients should be closely monitored for response. The Centers for Disease Control and Prevention (CDC) has published guidelines for the treatment of malaria in the United States (available at http://www.cdc.gov/malaria/).
All patients should be closely monitored for the possibility of hypoglycemia, which can occur either as a result of the infection itself or as an adverse effect of antimalarials, and is very common. Patients who are unable to eat should receive glucose-containing fluids as a continuous infusion. Patients treated with quinine should be monitored for findings suggestive of cinchonism, indicated by tinnitus and hearing loss which, if detected, should prompt dose reduction or change in therapy.
|Pharmacologic Agent(s)||Dosing Regimen|
|Quinine plus doxycycline||Quinine 10 mg/kg PO or IV q8hr for 7 days; doxycycline 100 mg PO q8hr for 7 days|
|Atovaquone-chloroguanide||Four adult tablets per day for 3 days (each adult tablet contains 250 mg atovaquone and 100 mg chloroguanide)|
|Artemether-lumefantrine||Four tablets initially, followed by 4 tablets 8 hr later, followed by 4 tablets twice daily on days 2 and 3 (each fixed-dose tablet contains 20 mg artemether and 120 mg lumefantrine)|
|Pyrimethamine-sulfadoxine||Two or 3 tablets PO once (each fixed-dose tablet contains 25 mg pyrimethamine and 500 mg sulfadoxine)|
|Artesunate plus mefloquine||
Artesunate 4 mg/kg/d for 3 days plus mefloquine 1250-mg single dose or 750 mg followed by 500 mg after 12 hr
|Artemether plus mefloquine||Artemether 3.2 mg/kg IM, first dose followed by 1.6 mg/kg after 12 hr, followed by 1.2 mg/kg/d for 3 days plus mefloquine 1250-mg single dose or 750 mg followed by 500 mg after 12 hr|
Cholera is a diarrheal disease caused by Vibrio cholerae. The disease is widely distributed in Asia, Africa, and South America. It occurs as an endemic form and has the potential to cause epidemics or pandemics periodically. An outbreak of cholera in Haiti in October 2010 resulted in more than 470,000 cases and 6,631 deaths over the next year.57
Of the more than 200 serogroups of V. cholerae, only 2–O1 and O139–cause clinical cholera. The natural reservoir of V. cholerae is water, and the predominant mode of spread is contamination of drinking water supplies. The bacterium can survive in a free-living state in water when conditions are favorable and reaches the intestinal tract of the host through ingestion of contaminated food or water. People who have decreased gastric acidity have a greater likelihood of becoming infected. Bacteria that survive the gastric environment and gain access to the small intestine cause disease by secreting an enterotoxin that stimulates the secretion of fluid and electrolytes into the lumen by the cells lining the small intestine.58 Excessive secretion leads to dehydration, and the dehydration resulting from cholera is the most severe of any infectious disease. The bacterium does not invade the wall of the small intestine.
Cholera manifests with an abrupt onset of watery diarrhea, soon followed by dehydration. Stools are thin and white, referred to as rice water stools. Vomiting is also common, and usually follows diarrhea. The severity of diarrhea and dehydration is variable, and it is important to realize that dehydration from cholera could be so severe that a previously healthy person could die of dehydration within hours of the onset of symptoms. In a much smaller proportion of patients, the presentation is abdominal distention rather than diarrhea because of accumulation of fluid within the intestinal lumen, a presentation known as cholera sicca. The most important complication is prerenal acute renal failure, and an incidence rate of 10.6 per 1000 was recognized in an outbreak in Peru in 1991.59
It is not difficult to diagnose a patient with cholera. Patients present with diarrhea and with surprising degrees of dehydration considering the short duration of the illness. For routine clinical care, microbiologic diagnosis cannot be made in a timely manner and is unnecessary.
The mainstay of therapy is adequate fluid resuscitation. Patients with mild dehydration can be treated with oral fluids. Patients who are more severely dehydrated, or those with moderate dehydration who are unable to tolerate oral fluids will require IV therapy. Rehydration should be provided in 2 phases, a rapid rehydration phase lasting 2 to 4 hours and a maintenance phase lasting the duration of the diarrhea. When possible, electrolytes should be monitored and abnormalities corrected. Since 1978, the World Health Organization oral rehydration solution (WHO ORS) has been recommended as the best choice for the management of diarrheal illnesses, including cholera.60 ORS is a sodium and glucose solution prepared by diluting 1 sachet of ORS in 1 L of safe drinking water. Effective and timely management will result is mortality rates < 1%. If ORS packets are not available, homemade solutions consisting of half a teaspoon of salt and 6 level teaspoons of sugar dissolved in 1 L of safe drinking water can be used.61 For IV rehydration, normal saline or Ringer’s lactate solution may be used. It must be noted that these fluids have low potassium content, and supplemental potassium may be required.
Antibiotics have been shown to reduce the duration of diarrhea and volume of stools.62 However, the use is of antibiotics in malaria patients is of secondary importance to hydration. A single 100-mg dose of doxycycline has been shown to be adequate antibiotic therapy.63 Alternative agents are trimethoprim-sulfamethoxazole, furazolidone, or erythromycin for 3 days, or a single dose of azithromycin, all of which can also be used in pregnant women and children.
The WHO ORS solution is hyperosmolar relative to plasma and concern has been raised that this solution may induce the development of an osmotically driven increase in stool output and resulting hypernatremia.64 Indeed, the use of this solution does not reduce stool volume or duration of diarrhea.65 In recent years, the use of reduced osmolarity ORS in areas endemic for cholera has gained favor,66 and this solution appears to be more effective than standard ORS.67 Regardless of the nature of the rehydration solution used, the primary principle of cholera treatment remains adequate rehydration.