Critical Care Medicine





Related conditions

Pulmonary alveolar proteinosis, infections in organ and hematologic transplant patients, pulmonary infections in HIV-infected patients, immunocompromised patient with fever and pulmonary infiltrates

1. Description of the problem

What every clinician should know

Nocardiosis should be suspected in any patient with cerebral, soft tissue or skin infections, especially in the setting of a concurrent or recent pulmonary process. Nocardiosis can present as suppurative disseminated disease or localized infection, and most commonly affects the lungs, central nervous system and skin. It occurs most often in immunocompromised patients who have cell-mediated immunodeficiency, specifically HIV patients, patients with organ or hematological transplants, patients taking chronic glucocorticoids and diabetics. However, one third of infected patients are immunocompetent, so Nocardia should not be disregarded when none of the classic risk factors are present.

Clinical features

Overview: Pulmonary disease is the most common presentation of nocardiosis, occurring in two-thirds of patients with Nocardia, and half of all pulmonary cases disseminate to extrapulmonary sites. Disseminated nocardiosis is defined as the presence of Nocardia in two non-contiguous sites, one of which is likely to be the lungs.

The most common site of dissemination is the brain due to a neural tropism, but Nocardia can disseminate to almost any organ, including the skin, lymphatic system, bones, heart valves, kidneys, joints, and the eye. When the primary focus cannot be identified, it can be assumed that the patient had transient pulmonary or cutaneous disease that resolved prior to presentation.

Pulmonary nocardiosis: Pulmonary nocardiosis can present acutely or subacutely with shortness of breath, dyspnea on exertion, cough, hemoptysis, or pleuritic chest pain. Symptoms can also be nonspecific, consisting only of fever, malaise, anorexia, weight loss, and night sweats. Pulmonary infection can extend to the mediastinum, causing mediastinitis and pericarditis, and to the pleura, causing an empyema.

Central nervous system (CNS) nocardiosis: CNS disease occurs in 20% of Nocardia cases in general, and up to 44% of disseminated cases. CNS nocardiosis most likely presents as an isolated parenchymal brain mass with focal neurologic deficits, fever, headache, or meningismus, although in immunocompetent patients CNS masses can go undiagnosed for several years. Nocardia can also cause meningitis, either in the presence or absence of a mass.

Bacteremia: Nocardia can be introduced into the bloodstream through a central venous catheter (and therefore can occur as a nosocomial infection). It can cause infection of pre-existing endovascular foreign bodies such as prosthetic valves or pacemakers, or it can occur in the blood during hematogenous dissemination from a primary site (typically the lungs). Up to 30% of the time, Nocardia bacteremia occurs in the presence of other bloodstream pathogens, most commonly gram-negative bacteria.

Cutaneous nocardiosis: Cutaneous nocardiosis can present as primary cutaneous disease, lymphocutaneous disease, secondary cutaneous disease disseminated from a primary focus or mycetoma.

Primary cutaneous disease is usually caused by inoculation of Nocardia through the skin, and can cause a subcutaneous nodule, pyogenic abscess, ulceration or cellulitis. Lymphocutaneous disease presents as a sporotrichoid process with nodules appearing in a linear fashion on the skin along the regional lymphatic drainage pathway. Cutaneous nocardiosis occurs in 2% of patients with disseminated disease; however, it is critical to treat all cutaneous disease as secondary disseminated disease until proven otherwise as primary and secondary cutaneous lesions cannot be differentiated by exam alone (see Figure 1).

Figure 1.

Several tender subcutaneous nodules on shins and thighs in a 53-year-old renal transplant patient with disseminated cutaneous nocardiosis.

In addition, Nocardia can cause infections in patients with a recent history of surgery or trauma, such as sternal wound infections. Finally, mycetoma are chronic infections caused by Nocardia (most commonly N. brasiliensis) or fungi. This occurs most commonly on the legs or feet after the organism is inoculated through the skin, forming a painless nodule that elicits a chronic inflammatory response, resulting in indurated lesions and sinus tract formation that can extend to deep tissues like muscle and bone.

Key management points

  1. Suspect Nocardia early in patients with cerebral, soft tissue or skin infections in the setting of a concurrent or recent pulmonary process. A red flag should go up in the setting of a patient with decreased cell-mediated immunity or malnutrition, such as HIV (most commonly with CD4 count under 100), post-transplant, chronic glucocorticoids (especially high-dose steroids), diabetes mellitus, malignancy (especially in the setting of recent chemotherapy or glucocorticoids). Other risk factors include a history of cytomegalovirus, high levels of calcineurin inhibitor use, use of tumor necrosis factor inhibitors, alcoholism, chronic lung disease (most commonly tuberculosis or pulmonary alveolar proteinosis) and chronic granulomatous disease.

  2. Obtain a culture as soon as possible whenNocardiais suspected. This may involve bronchoscopy with transbronchial biopsy in pulmonary nocardiosis, skin or soft tissue biopsy in the presence of cutaneous disease, lumbar puncture or brain biopsy in patients with isolated CNS disease, or blood cultures (although isolation of Nocardia in the bloodstream is rare). When an invasive culture is obtained, it grows Nocardia 85-90% of the time.

  3. Notify the laboratory when Nocardia is suspected, as identification can be difficult. Special stains and culture media are typically used and Nocardia can take up to 21 days to grow, so specimens must be held for longer than usual. Susceptibility testing should be performed on all clinically significant isolates, as susceptibility patterns can vary significantly between species.

  4. Have a low threshold to image the brain with computed tomography or preferably magnetic resonance imaging, since CNS dissemination is very common. Scan all patients with suspected Nocardia who also have symptoms of CNS disease, immunocompetent patients with pulmonary disease and all immunocompromised patients with pulmonary or cutaneous disease regardless of the presence of symptoms suggestive of CNS disease.

  5. Treat empirically with at least two antibiotics intravenously in severe disease, and then tailor antibiotic therapy once susceptibilities are available. In patients without CNS disease, empiric treatment should include trimethoprim-sulfamethoxazole (TMP-SMX) at 15 mg/kg IV daily, and amikacin. In patients with CNS disease, treat with TMP-SMX, a third-generation cephalosporin or imipenem, and possibly amikacin (in the setting of multi-organ disease). In the setting of TMP-SMX allergy, a third-generation cephalosporin or imipenem can be substituted.

  6. Surgical intervention is life-saving inNocardiapericarditis, which is almost always fatal in the absence of pericardial drainage. Surgical intervention also may be warranted in patients who have an empyema or mediastinal fluid collection, or in those who do not respond to initial antibiotic therapy.

2. Emergency Management

Stabilizing the patient

Patients with nocardial pulmonary disease, pericardial disease or sepsis can present acutely and be severely ill. Although patients present rarely with disease severe enough to warrant admission to the intensive care unit, severely ill patients with nocardiosis have an unusually high morbidity and mortality. In particular, in patients with pulmonary disease complicated by pericarditis, rapid pericardial drainage can be life-saving. Emergency management steps include:

  1. Rapid assessment of vital signs.

  2. Ensure oxygenation, including ventilatory support if warranted, in patients with pulmonary disease. In one study, many patients with pulmonary nocardiosis had respiratory failure with a mean PO2 of 55 mm Hg.

  3. If the patient shows evidence of tamponade physiology, obtain a stat echocardiogram and pursue pericardial drainage if a pericardial effusion is identified.

  4. If the patient shows septic physiology, obtain venous access and initiate fluid resuscitation.

  5. Obtain basic blood work (including CBC and chemistries), blood cultures, sputum cultures and chest imaging with pulmonary symptoms, brain imaging in certain situations as described above, and imaging of other symptomatic sites.

  6. Initiate broad-spectrum intravenous antibiotics, including antibiotics that cover Nocardia such as TMP-SMX, imipenem, a third-generation cephalosporin, and amikacin.

  7. Determine the level of monitoring needed (e.g. general ward versus intensive care unit). All patients with suspected or known nocardiosis, except those with isolated cutaneous infection, should be admitted to the hospital for at least several days to obtain a definitive diagnosis and initiate intravenous antibiotics.

3. Diagnosis

Diagnostic criteria and tests

The diagnosis of nocardiosis relies on isolation of Nocardia species in a clinical specimen, as there are no pathognomonic features that can definitively make the diagnosis of Nocardia in the absence of culture data. In addition, there are no widely accepted clinical diagnostic criteria for nocardiosis, as the clinical and radiographic findings are non-specific.

Invasive procedures should be considered early in the immunocompromised host, and the most common specimens from which Nocardia is isolated are respiratory secretions (sputum or bronchoalveolar lavage), skin biopsies or aspirates from deep collections. The clinician suspicious of the diagnosis of Nocardia should alert the laboratory if Nocardia is a possibility, as the cultures should be held up to 3 weeks.

Nocardia is not a typical human colonizer, so identification of the organism in a patient with disease that is clinically consistent with nocardiosis is usually sufficient to make the diagnosis. Furthermore, isolation of Nocardia in immunocompromised patients should almost always be considered as real disease.

However, transient colonization of sputum can occur in otherwise immunocompetent patients with underlying pulmonary pathology (such as cystic fibrosis, chronic obstructive pulmonary disease and bronchiectasis), which may not require specific therapy. In these patients, other disease processes should be ruled out; however, clinical suspicion of nocardiosis should nevertheless remain high when Nocardia is detected in one (or preferably two) respiratory tract specimens in the presence of clinical and radiographic findings consistent with nocardiosis, and in the absence of another disease etiology.

In the laboratory, Nocardia usually appears as long, branching, beaded gram-positive rods on direct smears from specimens. Nocardia species are both weakly acid-fast (they retain carbolfuchsin only if a weak decolorization procedure is performed) and partially acid-fast (often only a small proportion of cells retain the carbolfuchsin dye).

Although Nocardia will grow well on routine culture media, it grows better on buffered charcoal-yeast extract agar with antibiotics to suppress the normal flora. Colonies of Nocardia typically appear white and powdery and have a tan to orange reverse surface (Figure 2).

Figure 2.

Rough chalky-white colonies of Nocardia brasiliensis grown on Columbia blood agar.

The laboratory uses phenotypic tests to distinguish among the many species of Nocardia, and susceptibility testing should be done for all species. Of the over 30 species described, the most important causes of human infection are Nocardia asteroides sensu stricto, Nocardia farcinica, Nocardia nova, Nocardia brasiliensis, Nocardia pseudobrasiliensis, Nocardia otitidiscaviarum and Nocardia transvalensis. Newer commercially available identification systems such as Microscan RAI/HNID panels, the ID32C Yeast Identification System and the API 20C may allow for more rapid differentiation of Nocardia spp.

Normal lab values

There are no routine laboratory findings that are pathognomonic for Nocardia infections. Laboratory findings that may signal the presence of an underlying disease or process that would put patients at risk for nocardiosis should lead one to expect the diagnosis of Nocardia.

For example, laboratory evidence of a low CD4 count or other evidence of immunocompromise, such as hematologic malignancy or organ transplant, liver function abnormalities concerning for alcoholism or evidence of glucose intolerance consistent with diabetes mellitus would all raise suspicion for the possibility of Nocardia infection.

For patients with pulmonary nocardiosis, blood gas analysis will often reveal respiratory failure, with a mean PO2 of 55 mm Hg in one study. Radiographic manifestations of pulmonary nocardiosis include irregular nodules that can cavitate, reticulonodular or diffuse infiltrates, multiple abscesses, pleural effusions and progressive fibrosis in late disease (Figure 3).

Figure 3.

Large right-sided infiltrate with cavitation due to nocardia in a 6-year old boy with HIV.

The "halo sign" typically described in the setting of invasive aspergillosis has also been described with Nocardia infection. In HIV patients with CD4 counts less than 200/mm3, cavitary disease is particularly common, although radiographic findings may also show progressive alveolar infiltrates. Red flags as to the diagnosis of Nocardia include spread to contiguous structures, including pleura, pericardium, and soft tissue structures, often with fistula formation.

Cerebral disease can present in an insidious fashion, usually without meningeal signs. Cerebrospinal fluid obtained by lumbar puncture can sometimes demonstrate hypoglycorrhacia, but often it is normal. Therefore, brain imaging (MRI if possible) should be done when Nocardia is diagnosed, even in the absence of neurological signs.

Brain imaging typically reveals masses that are often mistaken for neoplasms. Therefore, a cerebral biopsy should be considered early in any patient in whom the diagnosis is suspected; in documented pulmonary nocardiosis a cerebral biopsy is not always necessary, although it should be noted that immunocompromised patients can have more than one concurrent disease process.

In disseminated disease (involvement of two noncontiguous organs), blood cultures should be obtained. However, blood cultures are positive in only a minority of cases, and patients with bacteremic nocardiosis are typically similar in presentation and risk factors to nonbacteremic patients with nocardiosis. Patients presenting with skin and subcutaneous lesions concerning for Nocardia should have a tissue biopsy to confirm the diagnosis.

Establishing the diagnosis

In an immunocompromised host or other host at risk for nocardiosis, growth of Nocardia in a tissue or sputum specimen should not be considered as a contaminant. Therefore, clinical or radiographic findings consistent with Nocardia in the setting of a positive laboratory finding are almost always diagnostic of clinically significant nocardiosis.

As mentioned above, even if Nocardia is isolated from one site such as the skin, disease at a second site such as lungs or brain should also be confirmed by biopsy in immunocompromised patients, even though this requires an invasive procedure, as it is possible for these patients to have more than one disease concurrently.

Other possible diagnoses

Pulmonary nocardiosis

The differential diagnosis of pulmonary nocardiosis includes other causes of pulmonary disease in an immunocompetent patient, including mycobacterial infections (M. tuberculosis or atypical mycobacterial pathogens), fungal infections (Aspergillus, mucormycosis, Cryptococcus neoformans), bacterial infections (including other pathogens found in patients with similar risk factors such as Rhodococcus equi, Pneumocystis jirovecii, Klebsiella pneumoniae, as well as more typical gram-positive and gram-negative pulmonary pathogens), viral infections, and lung cancer (either primary lung cancer or lung metastases). Mixed pulmonary infections can occur in immunocompromised patients.

CNS nocardiosis

The differential diagnosis of CNS nocardiosis includes other causes of brain mass (e.g. CNS lymphoma or metastases) or brain abscess (e.g. cryptococcoma, Aspergillus, Coccidioides immitis, mucormycosis, tuberculoma, bacterial abscess including anaerobic pathogens, and parasitic infections such as toxoplasmosis or cysticercosis). Although Nocardia often produces multiple brain abscesses, there is no typical presentation. Isolated Nocardia meningitis in the absence of a brain mass could also be mimicked by other causes of aseptic meningitis or, less commonly, as an acute bacterial meningitis.

Bacteremia and septicemia

The differential diagnosis of bacteremia and septicemia includes a wide variety of bacterial pathogens (either instead of or concurrent with Nocardia bacteremia) or fungal pathogens (e.g. Candida, Fusarium, etc.).

Cutaneous nocardiosis

The differential diagnosis of cutaneous nocardiosis includes nontuberculosis mycobacterial infections, fungal infections (e.g. Sporothrix schenckii, Aspergillus, Cryptococcus, Trichosporon beigelii, Fusarium, and Pseudallescheria boydii), sporotrichoid bacterial infections such as Erysipelothrix rhusiopathiae and Francisella tularensis, other routine gram-positive or gram-negative causes of cutaneous abscesses, including Staphylococcus aureus and Pseudomonas aeruginosa, and cutaneous parasitic infections such as leishmaniasis. Although mycetomas caused by Nocardia tend to evolve faster than those caused by fungi and other actinomycetes, they cannot be differentiated clinically.

Confirmatory tests

There are no specific tests that can confirm the diagnosis of Nocardia, other than identification of the bacteria in the laboratory. Serology is usually not clinically relevant, as there is not one serologic test that can detect all of the different species of Nocardia.

4. Specific Treatment

The optimal treatment agents and duration of therapy have not been well established in clinical trials as nocardiosis is not a common disease, so recommendations are based largely on animal models and expert opinion. Sulfonamides have long been the mainstay of treatment, but they should be administered in combination with other agents in immunocompromised patients or in patients with severe disease.

For severe or disseminated disease, at least two agents are recommended, with some experts recommending a three-drug regimen. Empiric therapy should be started when the diagnosis of nocardiosis has been made, and then later adjusted based on identification of the species and susceptibilites. Suggested initial treatment for the major nocardial syndromes are as follows:

Primary cutaneous nocardisosis


TMP-SMX and fluoroquinolone (addition of a fluoroquinolone should be considered for deep infections, especially mycetoma).

Pulmonary nocardiosis

TMP-SMX (should be used alone only in immunocompetent patients with mild disease).

TMP-SMX with ceftriaxone or moxifloxacin.

Imipenem and amikacin.

Disseminated nocardiosis

Imipenem and amikacin.

TMP-SMX, imipenem and amikacin.

TMP-SMX, ceftriaxone and amikacin.

CNS disease

TMP-SMX plus ceftriaxone, cefotaxime or imipenem. With multiorgan involvement, amikacin should be considered in addition to one of the above regimens.

Note: In patients with sulfonamide allergies or sulfonamide-resistant organisms, therapy should be started with amikacin plus one of the following: imipenem, ceftriaxone, cefotaxime.

Treatment should be started and continued intravenously for the first 3-6 weeks of treatment, and then switched to oral therapy if the patient is showing clinical improvement. Several factors may lead the clinician to alter these regimens:

1) Adverse drug effects: Common side effects of these medications include rash, bone marrow suppression, urinary lithiasis, renal toxicity and hypersensitivity. Thus, if these adverse drug effects arise, substitutions should be made accordingly. In addition to the regimens above, linezolid could replace any of the proposed drugs in the treatment of disseminated disease.

2) Drug resistance: Other potentially useful drugs that can be added or substituted when susceptibilities become available are other cephalosporins such ascefotaxime and cefepime, and other B-lactams such as amoxicillin-clavulanate or minocycline, and linezolid. The susceptibility patterns of the most common Nocardia species are as follows:

  • N. asteroides sensu stricto is typically susceptible to TMP-SMX, third-generation cephalosporins and amikacin. 64-98% of isolates are susceptible to imipenem.

  • N. farcinica is uniformly susceptible to amikacin but is resistant to other aminoglycosides such as tobramycin, as well as third-generation cephalosporins. Many isolates are susceptible to TMP-SMX and minocycline.

  • 97% of N. nova isolates are susceptible to TMP-SMX and virtually all are susceptible to third-generation cephalosporins, imipenem, and amikacin.

  • N. brasiliensis is typically susceptible to TMP-SMX and amikacin, and 88-100% of isolates are susceptible to third-generation cephalosporins. In contrast, only 20-30% of isolates are susceptible to imipenem.

  • Many N. transvalensis isolates are susceptible to TMP-SMX (88%), imipenem (90%), and third-generation cephalosporins (50%). N. transvalensis is usually resistant to amikacin and other aminoglycosides.

  • N. otitidiscaviarum is generally resistant to TMP-SMX but is usually susceptible to amikacin and minocycline.

3) Lack of clinical improvement: See below.

Drugs and dosages

Dosages of the drugs used to treat nocardiosis are as follows:

TMP 10 mg/kg/d divided into three daily doses for primary cutaneous nocardiosis.

TMP 10-20 mg/kg/d divided into three daily doses for pulmonary nocardiosis or disseminated disease.

Ciprofloxacin 500-750 mg po bid.

Ceftriaxone 2 g IV daily (q12h dosing in patients with CNS disease).

Cefotaxime 2 g IV q8h.

Moxifloxacin 400 mg po daily.

Imipenem 500 mg IV q6h.

Amikacin 15 mg/kg/d.

Refractory cases

If there is no clinical improvement after two weeks of therapy, the treatment regimen should be re-evaluated. Potential reasons for treatment failure include antibiotic resistance, poor penetration of drug into the diseased tissue compartment (such as aminoglycosides into CNS tissue), the need for surgical debulking, the presence of more than one disease process, or suboptimal absorption of oral medications. Therefore, management of the refractory case may include the following:

  1. Surgery may be particularly important for patients with brain abscesses, pericardial disease, mycetoma, empyemas, mediastinal collections and some ocular infections. In patients with thick-walled multiloculated abscesses, including patients with mycetomas, therapeutic aspiration may not be sufficient and more aggressive surgical debridement may be necessary. CNS disease that fails to respond to therapy may require aspiration or craniotomy; both were found to be equally effective in one series.

    Surgery should be considered when patients have not improved or have clinically deteriorated within the first 2 weeks of therapy, when lesions have not decreased in size on brain imaging within one month of therapy, and when abscesses are generally large or very accessible.

  2. For new or acquired drug resistance, or situations in which antibiotic penetration may be poor, antibiotic regimens can be altered to one of those listed above. Susceptibility testing should be done either by a reference laboratory or a local laboratory with experience in specialized susceptibility testing for Nocardia.

  3. Immunocompromised patients may fail therapy even if they are receiving antimicrobials to which their Nocardia species is susceptible, either due to overwhelming nocardial infection or the presence of other opportunistic infections. Immunosuppressive medications for transplant patients, or tumor necrosis factor inhibitors, should be stopped or tapered if possible.

  4. Measurement of serum drug levels should be considered for patients in whom there is concern about suboptimal absorption from the gastrointestinal tract, or in cases with poor therapeutic response despite seemingly appropriate antibiotic therapy. A serum sulfonamide level performed 2 hours after an oral dose at steady state can be used to confirm that GI absorption is adequate. The recommended therapeutic level of sulfonamide is between 100 and 150 mg/L.

5. Disease monitoring, follow-up and disposition

Expected response to treatment

Most patients show evidence of clinical improvement after 3-5 days, or at most 2 weeks. In cases in which appropriate antibiotic therapy is initiated promptly, outcomes are generally good even in the presence of disseminated disease. However, in patients with severe immunocompromise or in whom treatment is delayed, prognosis may be poor.

Duration of therapy depends on the site of disease, disease severity, clinical and radiographic response to therapy, and the patient's underlying level of immunosuppression. Primary cutaneous disease can often be treated for 1-3 months, although treatment of mycetoma can require many months. Pulmonary or disseminated disease is treated for at least 6 months when CNS disease is not present, and usually for 1 year or longer in patients with CNS involvement. All immunosuppressed patients, with or without CNS disease, should be treated for at least 1 year.

Patients should be seen at least monthly after hospital discharge, as it is important to monitor antibiotic doses and evidence of drug toxicity. Measuring antibiotic serum levels (such as sulfonamide levels) is recommended when patients are at risk of dose-related toxicity such as bone marrow suppression, in patients who require high doses of TMP-SMX, and in patients who have progressive or relapsing disease despite seemingly appropriate antibiotic therapy.

Sulfonamide intolerance may develop in a subset of patients being treated for nocardiosis, and this may be especially common in patients with HIV/AIDS. Patients who would benefit from continued TMP-SMX therapy can undergo desensitization, although severe reactions have been reported in non-HIV patients.

It is reasonable to obtain follow-up imaging at periodic intervals, such as 1, 6 and 12 months into the treatment course. Clinical deterioration or exacerbations can occur, as can drug toxicity, even during maintenance therapy, especially in immunosuppressed patients.

Incorrect diagnosis

Nocardia is not a typical contaminant, and therefore a diagnosis based on clinical findings consistent with nocardiosis in the setting of a positive laboratory identification is unlikely to represent an incorrect diagnosis. And, as mentioned above, patients with nocardiosis may not improve for a number of reasons despite having the correct diagnosis, especially when they are immunocompromised.

However, in cases where treatment is initiated empirically for nocardiosis (not recommended), either due to the riskiness of obtaining invasive tissue samples or negative cultures in the setting of recent or current antibiotic use, an incorrect diagnosis should be suspected when clinical improvement does not occur in the first two weeks, or if the patient continues to clinically deteriorate. In this case, every attempt should be made to obtain tissue for a definitive diagnosis.

It is also possible that more than one opportunistic infection or other disease process is present. For example, in a case in which pulmonary nocardiosis is improving while CNS disease is either not improving or progressing despite appropriate antibiotic therapy, it is possible that the CNS disease is attributable to another infectious or malignant process. In these settings, additional tissue should be obtained for examination.


Once patients have improved on 3-6 weeks of intravenous therapy, oral therapy can be instituted. In addition, the dose of TMP can be decreased from 10-20 mg/kg/d to 5-10 mg/kg/d divided into 2-4 doses in order to decrease the risk of adverse drug events.

Nocardiosis also can recur after treatment is stopped. In immunocompromised patients, such as HIV patients with persistently low CD4 counts, transplant patients on immunosuppressive therapy or patients maintained on steroid or cytotoxic therapy, secondary prophylaxis with an oral agent should be maintained until the net immune status of the patient improves. It is recommended that TMP-SMX prophylaxis be given on a daily basis, as twice or three times weekly TMP-SMX did not prevent recurrence of disease in bone marrow transplant patients.

When disease can be followed radiographically, such as pulmonary or CNS disease, imaging is recommended even after treatment is stopped. Some experts suggest repeating radiologic studies at six and twelve months after clinical cure.


More than 80 species within the genus Nocardia have been identified through techniques such as 16S ribosomal ribonucleic acid (rRNA) gene sequencing. The genus Nocardia is a member of the order Actinomycetales, which is a diverse group of organisms that share the same morphology characterized by branching filamentous gram-positive beaded rods that often fragment to coccobacillary forms. Other aerobic actinomycetes that cause human disease include Gordona, Tsukamurella, Streptomyces, Rhodococcus, Streptomycetes, Mycobacteria, and Corynebacteria. These organisms were once considered to be fungi because of their branching morphology, but analysis of their cell wall has led to their classification as bacteria.

The type and extent of disease caused by Nocardia depends on the route of infection, the tropism of the infecting species and the interaction between the host and organism. Although control of infection with Nocardia relies predominantly on the cell-mediated immune response, the initial response to Nocardia infection involves neutrophils and local macrophages. Early neutrophil mobilization serves to halt the infection while the cytotoxic T cell and macrophage response is mounted. The T cells, once primed, stimulate enhanced phagocytosis, direct toxicity to the bacteria and increased cellular response to the infection. Gamma delta T cells may be particularly important for the control of Nocardia infection.

Nocardia species have adapted ways to avoid both the host's innate and cell-mediated immune response. The bacteria demonstrate resistance to neutrophils through increased production of catalase and a cell surface-associated superoxide dismutase. Patients with specific defects in the oxidative burst, such as those with chronic granulomatous disease, are more vulnerable to infection.

Bacteria that are in log- or exponential-phase growth are resistant to phagocytosis by macrophages. For bacteria in a stationary (nongrowth) phase that are susceptible to phagocytosis, some nocardial strains can avoid hydrolysis by the host cell by inhibiting phagosome-lysosome fusion. Nocardia species that are able to inhibit phagosome-lysosome fusion within macrophages give rise to L-forms, which are cell wall-deficient variants. The presence of L-forms, which have been recovered from cerebrospinal fluid in patients with CNS nocardiosis and shown to demonstrate lifelong persistence in murine models, may explain the ability of nocardiosis to relapse and to recur years after successful initial antimicrobial therapy.

Once Nocardia enters susceptible cells within the lungs and airways or the skin, it initiates a suppurative reaction. Tissue specimens infected with Nocardia species typically show evidence of an acute pyogenic inflammatory reaction, with filamentous bacteria within the abscesses.

In fact, mycetoma can demonstrate bacterial macrocolonies called "sulfur granules" that are similar to those seen in actinomycosis. In the lungs, suppurative disease typically predominates, but granulomatous or mixed responses can occur, such as endobronchial inflammatory masses, pneumonia, pleural effusions and progressive fibrotic disease.

From any primary focus, Nocardia can spread hematogenously, via lymphatic spread to regional nodes or by direct spread to contiguous tissues, joints and bones. Patients with intact immune systems tend to have localized chronic disease, whereas hematogenous dissemination is much more common in immunosuppressed hosts.

Nocardia species are known for seeding the CNS. The type and extent of neurologic disease is determined by the neuroinvasiveness of the particular Nocardia species, which may in turn be determined by differences in expression of specific lectins in the host brain).

Members of the aerobic actinomycete group all share cell walls containing mesodiaminopimelic acid, arabinose, galactose and mycolic acids of differing chain lengths (responsible for variable acid-fastness). Nocardia species are also characterized by their ability to form hyphae above the surface of the substrate or medium, their ability to grow in lysozyme-containing media, and an ability to grow at 50C.


Nocardia is a ubiquitous environmental organism that can be found in soil, water and decaying organic matter, such as vegetable products. Nocardia is not a constituent of the human flora, so any isolate should be considered as a likely pathogen. However, transient colonization of sputum can occur in patients with underlying pulmonary pathology (such as cystic fibrosis, chronic obstructive pulmonary disease and bronchiectasis) and may not require specific therapy.

The most common route of entry is thought to be through inhalation of Nocardia that has become aerosolized on dust and other particles. Other portals of entry include ingestion of contaminated food and direct inoculation of contaminated soil or water through environmental sources (scratch, bite, other trauma) or nosocomial sources (peri-operative, catheter-related).

Nocardia cannot be transmitted directly from person to person. Therefore, respiratory or contact isolation of infected patients is not required. Although Nocardia is a well-recognized animal pathogen (with bovine mastitis as the most common veterinary disease), no cases of animal-to-human disease have been documented.

Globally, the most common cause of nocardial mycetoma in warm climates is N. brasiliensis (including the southern United States, Central and South America, and Australia) and the most common etiology of disseminated and respiratory infections is N. asteroides. Nocardia has been reported in every age and ethnic group, and tends to be 2-3 times more common in men for unclear reasons. It is possible that the male predominance is related to occupational risk - agricultural work appears to be an environmental risk factor for disease, with N. brasiliensis being the most likely pathogen - although some research has shown a possible protective effect of estrogen.

The incidence rate in the United States, last reported in 1970, is 500-1,000 new cases per year. However, the disease likely was and is underreported due to difficulty in diagnosis, and the incidence has almost definitely grown in the last 40 years as the population of susceptible patients (patients with HIV, organ transplants, hematologic malignancies and other forms of immunocompromise) has increased.

The biggest risk factor for nocardiosis is immunocompromise, typically involving deficiency in the cell-mediated response, although one third of infected patients are immunocompetent. The most common forms of immunocompromise are HIV, organ and hematopoietic stem cell transplant, malignancy and chronic glucocorticoid use. Other risk factors include pulmonary alveolar proteinosis, other forms of chronic lung disease such as cystic fibrosis and tuberculosis, chronic granulomatous disease, alcoholism, diabetes and chronic immunosuppressive medications such as glucocorticoids and tumor necrosis factor inhibitors.

Nosocomial nocardiosis has become a greater concern in the United States and other industrialized countries in recent years. Nosocomial transmission has been described in oncology and transplant units, where patients were exposed to infection through inhalation of contaminated dust and possible concurrent transmission through contaminated fomites or hands of staff. Hand transmission was also found to be a factor in sternotomy site infections due to N. farcinica. Finally, construction may have played a role in other documented postsurgical wound infections, and also in an outbreak on a liver unit.

In a review of 1,050 patients with nocardiosis, 64% were immunocompromised, with 17% having an underlying solid tumor or hematologic malignancy.

In organ transplant patients, the highest risk period is the first year after transplantation (although less likely in the first month after transplantation), presumably due to higher doses of immunosuppressant medications during this time. Glucocorticoid-sparing regimens result in lower rates of nocardial infection. Of note, the rates of infection were highest in heart and lung transplant patients, and most infected patients were being treated with prophylactic TMP-SMX.

Most organ transplant patients have been found to develop lung disease, but 20% had disseminated nocardiosis. Independent risk factors for development of nocardiosis in patients with organ transplants were high-dose glucocorticoids, cytomegalovirus infection in the past six months and use of high doses of calcineurin inhibitors in the past 30 days.

In patients with bone marrow transplants, the incidence of nocardiosis is approximately 340 times higher than among people in the general population.

In HIV patients, the incidence of nocardiosis is approximately 140 times higher than the incidence in the general population. Nocardiosis occurs in between 0.2-2% of patients, and most often in patients with extremely low CD4 counts (fewer than 100 cells/mm3). Although nocardiosis in HIV patients remains relatively rare, it is associated with high morbidity and mortality rates, and disease occurs more commonly in patients who are not taking TMP-SMX prophylaxis for the prevention of other opportunistic infections.

However, nocardiosis has also been diagnosed in patients who reported good compliance with TMP-SMX prophylaxis; this may be due to misdiagnosis with other pulmonary infections such as tuberculosis or bacterial pneumonia, concurrent infection with another opportunistic infection or as a manifestation of the immune reconstitution inflammatory syndrome.

Chronic glucocorticoid use leads to suppression of Th1-cellular immunity, and patients taking long-term steroids are at particularly high risk for disseminated or pulmonary disease.


Nocardia is characterized by its ability to disseminate to almost any organ and its tendency to relapse or progress despite appropriate therapy. Cure rates depend on the site and extent of disease, the patient's underlying immune status, the timeliness of initiation of appropriate therapy and the duration of therapy. For patients with limited skin or soft tissue disease, cure rates approach 100%. In contrast, the cure rate is estimated to be 90% in pleuropulmonary disease, 63% in disseminated infection, and 50% in patients with brain abscesses.

Treatment is often delayed in patients with nocardiosis, likely because the diagnosis is not usually obvious. In one study, the mean time to diagnosis of pulmonary nocardiosis was 42 ± 40 days, with a median of 30 days. For patients with disseminated disease, the mean time to diagnosis was 45 days, and in patients with CNS disease the mean time to diagnosis was 55 days.

Nocardiosis has a high mortality rate, reported as 7-44% in patients with disseminated infection. Mortality in immunosuppressed patients with disseminated disease has been reported as greater than 85%. Mortality in patients with brain abscesses is estimated at 31%, but is higher in the presence of multiple abscesses (41%) or underlying immunocompromise (55%).

Special considerations for nursing and allied health professionals.


What's the evidence?

Lederman, ER, Crum, NF. "A case series and focused review of nocardiosis: clinical and microbiologic aspects". Medicine. vol. 83. 2004. pp. 300-13.

(This is a case series of 5 patients with nocardiosis with varied presentations ranging from pneumonia and empyema to bacteremia to a sternal wound infection and primary cutaneous disease. This paper also includes a review of the literature from 1966 to 2003 that discusses epidemiology and microbiology.)

Lerner, PI. "Nocardiosis". Clin Infect Dis . vol. 22. 1996. pp. 891.

(This review article is a dated but still very relevant review of the classification and ecology of nocardiosis, methods of laboratory diagnosis, bacterial properties and the spectrum of disease presentations in immunocompromised and immunocompetent hosts. This article also reviews antimicrobial sensitivity testing and principles of antibiotic treatment.)

Beaman, BL, Beaman, L. "Nocardia species: host-parasite relationships". Clin Microbiol Rev . vol. 7. 1994. pp. 213-64.

(This article is a comprehensive review of the laboratory diagnosis, clinical aspects and mechanisms of pathogenesis of nocardiosis that includes an in-depth analysis of the immunology of the host-parasite relationship.)

Kontoyiannis, DP, Ruoff, K, Hooper, DC. "Nocardia bacteremia. Report of 4 cases and review of the literature". Medicine . vol. 77. 1998. pp. 255-67.

(This is a case series of four patients with Nocardia bacteremia and a review of the literature on the subject of Nocardia bacteremia. This review found that endovascular foreign bodies were a significant risk factor for bacteremia, described the characteristics of patients with bacteremia, and concluded that patients with bacteremia were otherwise virtually indistinguishable from patients without bacteremia.)

Martinez-Tomas, R, Menendez-Villanueva, R, Reyes-Calzada, S, Santos-Durantez, M, Valles-Tarazona, JM. "Pulmonary nocardiosis: Risk factors and outcomes". Respirology . vol. 12. 2007. pp. 394-400.

(This is a prospective observational study of patients with pulmonary nocardiosis at a hospital in Barcelona, Spain, over a 13-year period. During this time, 31 patients were identified with pulmonary nocardiosis, and the authors describe their characteristics, risk factors, mean time to diagnosis, treatment courses and outcomes.)

Sorrell, TC, Mitchell, DH, Iredell, JR, Mandell, GL, Bennet, JE, Dolin, R. "Nocardia Species". Principles and Practice of Infectious Diseases. Elsevier. 2005. pp. 2916-24.

(This is a very recent review in one of the most widely used infectious diseases texts that discusses the epidemiology, clinical presentation, diagnosis and treatment of a wide variety of Nocardia infections. Of special value, the authors review the differences in recommended antibiotic regimens for different Nocardia species.)

Agterof, MJ, van der Bruggen, T, Tersmette, M, ter Borg, EJ, van den Bosch, JMM. "Nocardiosis: a case series and a mini review of clinical and microbiological features". J Med . vol. 65. 2007. pp. 199-202.

(This is an in-depth discussion of one case of a Nocardia infection at a hospital in the Netherlands, supplemented by a review of eight other cases. The authors discuss the diversity of clinical presentations as well as updated principles of microbiologic identification.)

Ambrosioni, J, Lew, D, Garbino, J. "Nocardiosis: Updated clinical review and experience at a tertiary care". Infection . vol. 38. 2010. pp. 89-97.

(This is a case series of 28 patients with nocardiosis at a center in Geneva between 1989 and 2009 that discusses the risk factors, clinical features, diagnosis and management of disease in these patients. The authors also conduct a thorough review of the literature that addresses the challenges in diagnosis and novel molecular diagnostic tools.)

Wallace, RJ, Tsukamura, M, Brown, BA, Brown, J, Steingrube, VA. "Cefotaxime-resistant Nocardia asteroides strains are isolates of the controversial species Nocardia farcinica". J Clin Microbiol . vol. 28. 1990. pp. 2726-32.

(In this paper, the authors identify a specific pattern of resistance to cefotaxime and cefamandole and argue that the group of N. asteroides that manifest this resistance pattern are part of the taxonomic group N. Farcinica. The authors suggest that drug resistance patterns of N. asteroides may be associated with other specific taxonomic groups.)

Wallace, RJ, Brown, BA, Tsukamura, M, Brown, JM, Onyi, GO. "Clinical and laboratory features of Nocardia nova". J Clin Microbiol . vol. 29. 1991. pp. 2407-11.

(In this paper, the authors identify a specific pattern of resistance characterized by susceptibility to ampicillin and erythromycin that the authors feel is consistent with Nocardia nova. This is the first detailed description of N. nova as a human pathogen.)

Lee, GY, Daniel, RT, Brophy, BP, Reilly, PL. "Surgical treatment of nocardial brain abscesses". Neurosurgery . vol. 51. 2002. pp. 668-71.

(This is a case series of 11 patients treated for nocardial brain abscesses at the Royal Adelaide Hospital between 1970 and 2001. The authors review their clinical presentations, surgical treatment and outcomes. Their results suggest that aspiration alone is safe and efficacious treatment for most patients with nocardial brain abscesses.)

McNeil, MM, Brown, JM. "The medically important aerobic actinomycetes: Epidemiology and microbiology". Clin Microbiol Rev . vol. 7. 1994. pp. 357-417.

(This is a review of the epidemiology and microbiology of the most important human aerobic actinomycetes, including Nocardia. The authors discuss the challenges of identifying new pathogenic species of Nocardia, as well as clinical and laboratory techniques to assess antimicrobial susceptibility and adequacy of treatment.)

Choucino, C, Goodman, SA, Greer, JP. "Nocardial infections in bone marrow transplant recipients". Clin Infect Dis . vol. 23. 1996. pp. 1012-9.

(This is a review of six cases of nocardiosis occurring in a population of bone marrow transplant patients at Vanderbilt University Hospital, as well as four cases reported previously in the literature. In this review, most patients had extensive exposure to soil or dust prior to infection and mortality rates were high, demonstrating that Nocardia is an important although rare complication of bone marrow transplantation.)

King, DP, Hyde, DM, Jackson, KA. "Cutting edge: Protective response to pulmonary injury requires gamma delta T lymphocytes". J Immunol . vol. 162. 1999. pp. 5033.

(The authors of this paper analyze the importance of gamma delta T cells in the immunologic response to Nocardia infection using the mouse model. Their data demonstrate that gamma delta intraepithelial lymphocytes can protect the host by modulating the immune response after nocardial infection.)

Beaman, BL, Maslan, S. "Virulence of Nocardia asteroides during its growth cycle". Infect Immun . vol. 20. 1978. pp. 290-5.

(The authors use a mouse model to study the mechanisms Nocardia species have developed to avoid both the host's innate and cell-mediated immune response. Specifically, they discuss how different stages of growth affect virulence.)

Kuipers, S, Aerts, PC, van Dijk, H. "Differential microorganism-induced mannose-binding lectin activation". FEMS Immunol Med Microbiol . vol. 36. 2003. pp. 33-9.

(The type and extent of neurologic disease is determined by the neuroinvasiveness of the particular Nocardia species. The authors of this paper used blood from healthy people to investigate concentrations at which Nocardia species activate the mannose-binding lectin pathway and determined that Nocardia activated this pathway well.)

Brown, JM, McNeil, MM, Murray, PR, Baron, EF, Jorgensen, JH. "Nocardia, Rhodococcus, Gordona, Actinomadura, Streptomyces, and other aerobic actinomycetes". Manual of Clinical Microbiology. ASM Press. 2003. pp. 502-31.

(This chapter on Nocardia from the well-known reference volume in the field of clinical microbiology discusses the laboratory diagnosis and therapeutic testing of Nocardia species.)

Hernandez-Hernandez, F, Lopez-Martinez, R, Mendez-Tovar, LJ, Manzano-Gayosso, P. "Nocardia brasiliensis: in vitro and in vivo growth response to steroid sex hormone". Mycopathologia. vol. 132. 1995. pp. 79.

(These in vitro experiments provide important information about the effect of hormones on Nocardia species, in an attempt to better understand the lower rates of actinomycetoma in women compared to men. The authors of this paper grew five N. brasiliensis strains on agar containing different concentrations of estradiol, progesterone, and testosterone, and measured colony diameters weekly for 7 weeks.)

Houang, ET, Lovett, IS, Thompson, FD. "Nocardia asteroides infection - A transmissible disease". J Hosp Infect. 1980. pp. 31-40.

(The authors describe an outbreak of N. asteroides in a renal intensive care unit in predominately kidney transplant patients, and suggest that nosocomial transmission occurred through inhalation of dust and air inside the unit. The index patient had been excreting Nocardia in urine from her nephrostomy tube.)

Wenger, PN, Brown, JM, McNeil, MM, Jarvis, WR. "Nocardia farcinica sternotomy site injections in patients following open heart surgery". J Infect Dis . vol. 178. 1998. pp. 1539-43.

(This is a case-control study comparing five patients with N. farcinica sternotomy site infections after open heart surgery to five patients with open heart surgery who did not develop sternotomy site infections. This study, the first study to describe a nosocomial N. farcinica outbreak, contributed to our ability to understand the epidemiology and microbiology of nosocomial person-to-person spread of this pathogen.)

Sahathevan, M, Harvey, FA, Forbes, G, O'Grady, J, Gimson, A. "Epidemiology, bacteriology and control of an outbreak of Nocardia asteroides infection on a liver unit". J Hosp Infect. vol. 18. 1991. pp. 473-80.

(This paper describes a common-source outbreak of N. asteroides in seven patients with liver disease, and identified the most probable source of the outbreak to be contaminated brick and plaster dust from building work in an area near the liver ward. The authors discuss infection control measures critical for controlling this outbreak, including closing the ward, thorough cleaning and formaldehyde fumigation.)

Smego, RA, Moeller, MB, Gallis, HA. "Trimethoprim-sulfamethoxazole therapy for Nocardia infections". Arch Intern Med . vol. 143. 1983. pp. 711-78.

(This case series describes 19 patients with Nocardia infections seen at Duke University Medical Center treated with trimethoprim-sulfamethoxazole either alone or in combination with other antibiotics or surgery. This case series demonstrates the efficacy of TMP-SMX as first-line therapy for Nocardia infections.)

Corti, ME, Villafane-Fioti, MF. "Nocardiosis: a review". Int J Infect Dis . vol. 7. 2003. pp. 243-50.

(This is a review of the epidemiology, pathophysiology, clinical presentation, diagnosis, management and prognosis of Nocardia infections by authors from Argentina. The authors discuss the high mortality rates of Nocardia infections in immunocompromised hosts and the importance of prompt diagnosis and treatment.)

Mamelak, AN, Obana, WG, Flaherty, JF, Rosenblum, ML. "Nocardial brain abscess: treatment strategies and factors influencing outcome". Neurosurgery . vol. 35. 1994. pp. 622-31.

(This is a case series of 11 patients with nocardial brain abscess treated between 1971 and 1993, as well as a review of the 120 documented cases in the literature between 1950 and publication of this article. This paper reviews the clinical findings, patient characteristics, mortality rate, recommended diagnostic modalities of Nocardia brain abscesses, and antibiotic treatment regimens for these patients.)

Related Resources

You must be a registered member of Cancer Therapy Advisor to post a comment.

Regimen and Drug Listings


Bone Cancer Regimens Drugs
Brain Cancer Regimens Drugs
Breast Cancer Regimens Drugs
Endocrine Cancer Regimens Drugs
Gastrointestinal Cancer Regimens Drugs
Gynecologic Cancer Regimens Drugs
Head and Neck Cancer Regimens Drugs
Hematologic Cancer Regimens Drugs
Lung Cancer Regimens Drugs
Other Cancers Regimens
Prostate Cancer Regimens Drugs
Rare Cancers Regimens
Renal Cell Carcinoma Regimens Drugs
Skin Cancer Regimens Drugs
Urologic Cancers Regimens Drugs

Sign Up for Free e-newsletters