Hematology

Chronic granulomatous disease

Chronic granulomatous disease

What every physician needs to know:

Background

Chronic granulomatous disease (CGD) is a deficiency of microbicidal oxidant production affecting neutrophils, eosinophils, monocytes, and some tissue macrophages. There are five genetic forms of CGD affecting different subunits of the phagocyte NADPH (nicotinamide adenine dinucleotide phosphate) oxidase (gp91phox, p47phox, p22phox, p67phox, p40phox). X-linked CGD caused by mutations in gp91phox accounts for almost 65 to 75% of cases. Mutations in p47phox are responsible for most autosomal recessive cases of CGD, accounting for about 20 to 25% of all patients with CGD, affecting males and females equally.

Clinical presentation

Patients with CGD present with recurrent bacterial and fungal infections that most commonly include pneumonia, lymph node infection, and deep soft-tissue abscesses. Most patients will appear quite well between infections, with severe infections that may occur at intervals of several months to several years, leading to delay in consideration of the diagnosis of CGD. Most often the diagnosis of CGD is first considered because an infection persists, not responding as expected to initial therapy. Liver abscess, osteomyelitis, brain abscess, and other deep organ infections may also occur.

CDG-related infections

CGD-related infections occur with a characteristic array of microorganisms (typically Staphylococcus aureus, Serratia marcescens, Burkholderia cepacia and other burkholderia species, nocardia species, Aspergillus fungal species, as well as dematiaceous molds). CGD patients are not particularly susceptible to Candida albicans, though there is increased susceptibility to Candida glabrata (torulopsis).

In areas endemic for tuberculosis, patients with CGD are particularly susceptible. In the United States and other areas not endemic for tuberculosis, CGD patients are often misdiagnosed as having tuberculosis because of the microscopic granulomas seen in the histology of biopsies of CGD patient tissues infected with other organisms. These granulomas may resemble the caseating granulomas of tuberculosis (but lack detectable acid fast bacteria and do not grow tuberculosis).

Rarely, CGD patients can have infections with atypical mycobacteria, but generally if a patient presents with widespread systemic atypical mycobacteria infection, one should look for other severe immune deficiencies rather than CGD. Clostridium difficile infections of the bowel are common in CGD patients and can be difficult to eradicate, with relapses common. CGD patients are not susceptible to infection with commensal organisms such as Escherichia coli. They handle viral infections in a manner similar to normal hosts, though there can be an increased incidence of secondary bacterial infections following viral infection. For this reason, CGD patients can safely be given live virus vaccines.

CGD patients should receive all childhood vaccinations including live virus vaccines, but should never receive Bacillus Calmette-Guerin (BCG) vaccine for tuberculosis, that is standard tuberculosis prophylaxis in most of the world (though not in the United States). BCG vaccination in infants and young children with CGD can cause an often fatal, difficult to treat systemic BCG infection.

CGD may vary significantly in severity, dependent upon the genotype and type of mutation. While many patients present with characteristic findings and are diagnosed in the first decade of life, some patients with partial oxidase function may have such infrequent infections, that the diagnosis of CGD is not considered until the patient presents with an infection typical of CGD in adulthood.

Inflammation and granuloma formation

CGD is also associated with excessive inflammation and granuloma formation which, in almost 20% of patients, affects the gastrointestinal (GI) tract. This may manifest as a Crohn's disease-like colitis that can include diarrhea and/or chronic rectal fissures and fistulas. In some patients it may be the colitis, rather than recurrent infections, that is the dominant problem. There are cases reported of individuals carrying a diagnosis of childhood Crohn’s disease for many years, until they present with a fungal pneumonia or other infection that suggests a diagnosis of CGD. This is not common enough to increase the incidence of CGD in patients with Crohn’s disease above that of the incidence in the general population, and routine testing for CGD in patients with Crohn’s disease is not recommended.

While mild colitis in CGD is often remarkably responsive to low doses of alternate-day prednisone, more severe chronic colitis may require prolonged daily higher dose prednisone and other modalities of established treatments for Crohn’s disease. More detailed discussion of management of severe colitis in CGD will be discussed in the later sections, but it should be noted that use of anti-tumor necrosis factor (TNF) biologics to treat severe colitis in CGD patients is controversial, because anti-TNF agents significantly increase susceptibility of CGD patients to life-threatening infections.

It is important to note that sporadic macroscopic sterile granulomas of the GI tract or genitourinary (GU) tract also may occur in CGD patients in the absence of colitis. Occasionally, granuloma-related pyloric obstruction with vomiting may be the first presenting symptom of CGD in early infancy. In older children with CGD, unexplained sporadic postprandial vomiting, or chronic postprandial abdominal discomfort, can be the presenting symptom of granuloma related partial pyloric obstruction.

Boys with CGD can present with sporadic sudden bladder pain, or pain on micturition with or without impairment of stream, from sterile granuloma partially obstructing the bladder exit, or from granulomatous bladder wall thickening. All of these localized obstructive granulomatous processes in CGD are very responsive to a short burst of high-dose prednisone followed by a rapid taper, although some patients may require prolonged or permanent low-dose alternate-day prednisone to prevent recurrence of obstruction.

Autoimmune disorders

In addition to colitis or other chronic inflammation of the GI tract, which affect more than 10% of CGD patients, and the sporadic obstructive granulomas, CGD patients also have a greater than expected incidence of autoimmune disorders that include discoid or systemic lupus, sarcoidosis, Behçet's syndrome, idiopathic juvenile rheumatoid arthritis, autoimmune thrombocytopenia, anti-phospholipid clotting disorder, Ig (immunoglobulin) A nephropathy, idiopathic autoimmune pericardial effusion, progressive idiopathic pulmonary inflammation, and other poorly defined autoimmune syndromes.

Bone marrow transplant for CGD

Bone marrow transplantation can result in permanent cure of CGD, but there is risk of acute and chronic graft-versus-host disease, risk of life-threatening transplant-related infections, and risk from other complications of transplantation that make risk/benefit decisions difficult. There is no clear consensus regarding which CGD patients should be referred for transplantation but there are some emerging guidelines, and a more detailed discussion of this issue is provided in the sections to follow.

Are you sure your patient has chronic granulomatous disease? What should you expect to find?

Most patients with CGD lack any characteristic findings on physical examination, unless there is ongoing infection or inflammation with signs and symptoms localized to the site of infection or inflammation. Children with CGD can have either generalized or localized lymphadenopathy without infection, but if a lymph node is warm or tender to pressure on physical exam, an infection is likely. Young children with CGD can also have modest to significant splenomegaly and hepatomegaly without infection.

Family history and the patient’s medical history and current symptoms are most important to considering a diagnosis of CGD. A family history of unexplained male deaths from infection on the maternal side can suggest X-linked immune deficiency that includes X-linked CGD. A history of recurrent pneumonias and/or lymph node infections that persist, despite usual standard-of-care empiric management should trigger consideration of testing for CGD.

CGD is a likely diagnosis in patients presenting with staphylococcal, Serratia, or fungal liver abscess. A patient presenting with a pneumonia from which aspergillus or other hyphal mold, burkhoderia, nocardia, or Serratia is detected is likely to have CGD. Serratia is a particularly common cause of infection in infancy in CGD. In summary, one should suspect CGD in a patient who has a pneumonia that persists despite prolonged antibiotic therapy, in any patient with an unexplained liver abscess, and in any patient, particularly an infant or young child with serratia, soft-tissue abscess, or osteomyelitis.

The definitive diagnosis of CGD can only be made using one of the assays that measure neutrophil oxidant production. The classic assay is the nitroblue tetrazolium dye assay (NBT assay), but in the last 15 years this has been increasingly replaced by the dihydrorhodamine (DHR) flow cytometry assay. The DHR assay provides more diagnostic information and is a more sensitive assay for detecting partial-function CGD variants, which can be missed when using the NBT assay.

Beware of other conditions that can mimic chronic granulomatous disease:

Oxidase function assays can be spuriously low in patients who are critically ill and hemodynamically unstable from severe trauma, burns, or acute infection. For this reason, results from laboratory testing for CGD in that setting should be considered only “presumptive” and must be repeated at a time when the patient is not critically ill. However, infection, per se, does not adversely affect laboratory testing for CGD. Several reports suggest that patients with severe deficiency of glucose-6-phosphate dehydrogenase have low activity of neutrophil oxidase and increased susceptibility to infections.

Beware of the misdiagnosis of Aspergillus hypersensitivity pneumonitis in an adolescent or young adult presenting with the symptom complex of cough, severe hypoxia, mild to moderate diffuse pulmonary miliary/reticular infiltrate on computerized tomography x-ray exam, and Aspergillus seen or grown from sputum or from bronchoscopy lavage. Rarely, this can be the presenting diagnosis in an adolescent or young adult patient who is not previously known to have CGD, but who actually has a mild variant form of CGD and who has had an intense exposure to Aspergillus spores from spreading mulch or handling manure.

Such CGD patients actually have a diffuse inhalation pneumonia with Aspergillus that requires both intensive and usually very prolonged treatment with antifungal agents with activity against Aspergillus and other molds, plus short-term treatment with high-dose steroids to prevent early hypoxic death from CGD-related hyperinflammatory response in the lungs. If a misdiagnosis of Aspergillus hypersensitivity pneumonitis is made in a CGD patient presenting with diffuse inhalation-Aspergillus pneumonia, the use of steroids alone without intensive treatment with antifungal antibiotics can lead to a fatal outcome.

A great many other immune deficiencies other than CGD may be associated with susceptibility to recurrent pyogenic infections. For example, patients with IRAK4 (interleukin-1 receptor-associated kinase 4) deficiency, with STAT (Signal Transducers and Activators of Transcription)-1 mutations or with STAT3 deficiency (Job's syndrome, autosomal dominant hyperimmunoglobulin E recurrent infection syndrome) have recurrent infections with S. aureus. So these and other immune deficiencies must be investigated and ruled out if the DHR or NBT test is not diagnostic of CGD.

Which individuals are most at risk for developing chronic granulomatous disease:

Demographic differences

Published demographic studies of CGD conducted in several countries, demonstrate that the incidence of CGD is similar across all races and ethnic groups at approximately 1 in 150,000 to 200,000 births. Additionally, the X-linked form of CGD comprises about 65 to 75% of cases in all races and ethnic groups. For this reason 80 to 85% of all CGD patients are males and about 15% are female. However, in population groups where consanguinity is more common, the relative incidence of autosomal recessive forms of CGD is higher, and founder effects can lead to clusters of one or another type of the autosomal recessive forms of CGD in a particular consanguinous community.

Genetic predispositions

The p47phox-deficient autosomal recessive form of CGD is significantly more common than the other three autosomal recessive forms of CGD combined, accounting for 20 to 25% of all cases of CGD. The gene encoding p47phox, NCF1, has two pseudogenes, likely leading to a high incidence of gene conversion (mis-crossover event between either of the two pseudogenes and NCF1) leading to inactivation of the normal gene.

Formal genetic counseling is an important element of care for the patient and the patient’s family, once a diagnosis of CGD is confirmed by laboratory tests of oxidase activity. More detailed discussion of laboratory studies including genetic sequencing for the diagnosis of CGD will be provided in the next section.

In general, the X-linked form of CGD affects males, and subsequent male births to a carrier mother of a patient with X-linked CGD have a 50% chance of having inherited CGD. Female carriers of X-linked CGD have a mixture of oxidase negative (CGD phenotype) and oxidase normal neutrophils in their blood, because in females one of the X-chromosomes inactivates in a random fashion during embryogenesis, leading to a mixture of hematopoetic stem cell clones working from one or the other active X-chromosome. This gives rise to oxidase-negative and oxidase-normal neutrophils.

The ratio of normal and CGD-like neutrophils in female carriers of X-linked CGD, once set by chance early in embryogenesis, tends to remain the same through life. However, it is important to test and know this ratio in female carriers, because those with less than 10% oxidase-normal cells are at increased risk of both CGD-related infections and the inflammation/autoimmune problems associated with CGD.

There are reports of rare female carriers with extreme skewing of X-chromosome inactivation, resulting in less than 5% oxidase-normal neutrophils, who can have the full clinical spectrum of problems associated with CGD. Furthermore, female carriers of X-linked CGD with only modestly low levels of oxidase-normal neutrophils (20 to 30%) who have been clinically healthy all their lives, may in their seventh or eighth decade of life present with an infection characteristic of CGD.

It may be that as immune function wanes in the elderly, the 20 to 30% oxidase-normal neutrophils that was adequate for host defense in youth becomes inadequate much later in life, where overall immune function has decreased. As previously noted, autosomal recessive forms of CGD affect males and females equally. Both the mother and father must be genetic carriers of a mutation in the same autosome CGD gene, and the risk of subsequent children inheriting autosomal recessive CGD in that setting is 25%. Neutrophil oxidase function is normal in all neutrophils of male or female carriers of any of the autosomal recessive forms of CGD.

What laboratory studies should you order to help make the diagnosis and how should you interpret the results?

Dihydrorhodamine flow cytometry test

The definitive test currently recommended for CGD is the dihydrorhodamine (DHR) flow cytometry test of neutrophil oxidase activity (DHR assay). The DHR assay requires a minimum of 2 to 3ml of anticoagulated blood. In many major medical centers, the clinical pathology laboratory is capable of performing this flow cytometry assay. The "neutrophil oxidase burst assay" is also available from such commercial laboratories as ARUP Laboratories, The Mayo Clinical Laboratories, Focus Diagnostics and others.

In general, a blood sample from the patient, the patient’s mother, and a normal control (often the patient’s father) should all be sent for analysis together, to get the most information from the assay and to control for any shipping or other artifacts. In order to properly understand how to interpret the results provided to the requesting physician from the reference lab, it is important to have some understanding of how the assay is performed.

In brief, the assay is performed by lysing the red cells in the blood sample, allowing the leukocytes to take up DHR, and then treating the cells with the potent non-specific stimulant, phorbol myristate acetate, which maximally activates the neutrophil NADPH oxidase over 15 minutes. For each sample, a control is also run that has not been stimulated with phorbol myristate acetate. As hydrogen peroxide and other oxidants are generated in the stimulated sample, the DHR is oxidized, converting it to a highly fluorescent product, increasing the fluorescence intensity of normal neutrophils by more than 100-fold.

Flow cytometry is performed with gating on the cluster of events corresponding to where neutrophils localize on a forward-scatter/side-scatter dot plot. Generally the data is shown as a histogram curve with the Y-axis (vertical) plotting number of cells and the X-axis (horizontal) plotting fluorescence intensity.

If there has not been any artifactual activation of the neutrophils prior to analysis from mishandling, the unstimulated neutrophils from the patient, the mother, and the control should appear very similar in each histogram as a single sharp peak at about a mean fluorescence intensity (MFI) on the X-axis at about 50 to 115 fluorescence units (FU). If oxidase activity is normal in the activated samples, there should be a single sharp peak at about an MFI of 5,000 to 11,500FU, representing a more than 100-fold increase in fluorescence.

CGD patients with function-null mutations whose neutrophils have no residual neutrophil oxidase activity may show little or no change from unstimulated baseline. However, CGD patients who have partial-function mutations, allowing some residual neutrophil oxidase activity, may demonstrate some increase in fluorescence with the histogram, showing a single broad-based peak between an MFI of 225 and 1,500, corresponding to a 3- to 35-fold increase in fluorescence.

As noted in the section above, female carriers of X-linked CGD are functional mosaics, in that they have a mixture of oxidase-negative CGD phenotype neutrophils and oxidase-normal neutrophils in their peripheral blood. The DHR-assay histogram of stimulated neutrophils from a female carrier of X-linked CGD will have two distinct peaks.

This provides important information regarding the genetic subtype of CGD. For example, a male with absent or low oxidase activity on the DHR assay diagnostic for CGD, whose mother’s DHR assay demonstrates the two peaks diagnostic of being an X-linked carrier, has the X-linked form of CGD. If DHR assay of the patient’s mother’s neutrophils show only a single peak of oxidase-normal neutrophils, then it is strong evidence that the patient likely has one of the autosomal recessive forms of CGD.

Recent studies have shown that the amount of residual oxidase activity produced by neutrophils, as measured in the DHR assay, is significantly correlated with survival. Rarely, X-linked CGD may occur in a male child because of a mutation first arising in the affected patient and not present in the mother. In that case, the mother’s DHR assay may not demonstrate two peaks, leading to the reasonable but erroneous assumption that the type of CGD is autosomal. This situation of spontaneously occurring new mutation in the X-linked CGD gene occurs in about 2 to 3% of newly identified X-linked CGD cases with no prior family history.

Genetic tests and protein assays

Although it is not critical for day-to-day care and management to know the genetic subtype of CGD or the specific genomic mutation responsible for causing CGD in a patient, it can be important for providing counseling regarding prognosis, for genetic counseling of other family members, and for family planning.

Commercial sequencing is available from Gene Dx, Inc. for the four genes most commonly affected in CGD (CYBB,CYBA, NCF1, NCF2). Because only two cases of CGD caused by p40phox deficiency are known, commercial sequencing of NCF4 is not available. Since gene sequencing can be expensive, it is helpful to narrow down the field. If the mother of the patient has two peaks on the DHR assay indicating X-linked carrier status, then the CYBB gene encoding gp91phox should be sequenced in the patient. If the mother does not appear to be a carrier of X-linked CGD in the DHR assay, then the common GT-deletion splice-junction mutation at exon 2 in NCF1 (encoding p47phox) should be looked for, as this is the next most common form of CGD.

Specialized centers, focused on the study of CGD patients, such as the National Institutes of Health may be able help narrow down the probable CGD subtype by performing Western-blot assays to determine which oxidase subunit is absent from neutrophils of the patient. This may determine which oxidase subunit gene should be sequenced.

What imaging studies (if any) will be helpful in making or excluding the diagnosis of chronic granulomatous disease?

There are no imaging studies that are diagnostic for CGD. However, judicious use of imaging is critical to the long-term management of CGD, and for the diagnosis and treatment of infections

Computerized axial tomograph scan

Particular note should be made of the fact that pneumonia, particularly fungal pneumonia, is the most common potentially life-threatening infection occurring in CGD patients. Early detection of lung infection is critical to the long-term management of CGD, particularly because CGD patients with early fungal pneumonia can be devoid of any signs or symptoms, in that fever, cough or malaise may be absent, and all blood tests may be normal including blood count, erythrocyte-sedimentation rate or C-reactive protein. The small infiltrates associated with early infection are often undetectable on a regular chest X-ray, but often are easily seen on a computerized tomography (CT) scan of the chest.

In CGD patients, a yearly routine-screening CT scan of the chest is cost-effective with respect to picking up unappreciated, very slowly advancing fungus pneumonia. Also, unless a CGD patient has signs and symptoms of a routine upper respiratory tract viral infection, a CT scan of the chest is appropriate for any patient presenting with any constellation of persistent signs and symptoms that could be related to pneumonia. These include fever, malaise, cough, shortness of breath, and/or chest pain lasting more than 3 or 4 days.

Because of the reasonable concern for excessive diagnostic radiation exposure, particularly in the pediatric population, it is recommended that injection of radiologic contrast dye not be used for most diagnostic chest CTs in CGD patients, as it is not needed to detect a pneumonia, though may be useful if mediastinal- or chest-wall infection is suspected.

In addition, many radiologic services are exploring reduced radiation algorithms for chest CTs and finding that surprisingly low doses of radiation may be used without compromising resolution of the CT. This can be particularly important in children with CGD who are likely to gain a substantial lifetime exposure to radiological examinations. It is important to ask your radiologist if the CT of the chest in a CGD patient can be done with a low-radiation/no-contrast-dye algorithm.

If you decide the patient has chronic granulomatous disease, what therapies should you initiate immediately?

Antibacterial/antifungal oral antibiotics

Universal standard of practice for long term management of CGD includes a daily regimen of combined antibacterial plus antifungal oral antibiotic infection prophylaxis. Less universal and more controversial, as will be discussed further below, is three times weekly injections with human recombinant interferon gamma. Specific prophylaxis recommendation with dosing that may begin as early as the third month of life includes:

  • Daily oral trimethoprim-sulfamethoxazole (clotrimazole) about 4.6mg/kg/day trimethoprim – 22.8mg/kg/day sulfamethoxazole

- Since the half-life in the blood is relatively long, the dosing can be given once a day or for convenience, in a divided dose, morning and evening. For very young children, given the liquid form, exact dosing is possible, but for older children and adults given tablets, dosing may be adjusted upward as needed to accommodate the single-strength or double-strength tablet formulations, given either all at once or in a divided twice daily dose.

  • Daily oral itraconazole at about 4 to 5mg/kg/day

- Again, since half-life in the blood is relatively long, the dosing can be given once a day. For very young children, given the liquid form, exact dosing is possible, but for older children and adults given tablets, dosing may be adjusted upward as needed to accommodate the available dosing size.

  • Three times weekly subcutaneous injections of human recombinant interferon gamma

- 50 mcg/m2 surface area, with special dosing adjustment for infants under 10kg.

Because CGD patients are not at risk of developing infection from their normal commensal organisms in the colon or on the skin, antibiotic prophylaxis is directed against developing infection from environmental pathogens most likely to cause infection in CGD patients. For bacterial prophylaxis trimethoprim-sulfamethoxazole is active against the key risk bacterial CGD pathogens, while itraconazole is active against the key risk mold CGD pathogens. Both agents have a relatively low profile of side-effect risk, including a relatively low effect in altering the normal host commensal microbial ecology.

Occasionally young children may develop neutropenia from clotrimazole, and for this reason, a complete blood count at baseline and then periodically during the first year of prophylaxis in young children is recommended. Often the neutropenia can be ameliorated without affecting antibacterial potency, by treating with daily leukovorin (not folate); 1 to 5mg per day as needed to ameliorate the clotrimazole-induced neutropenia.

If leukovorin does not correct the anemia, or in patients allergic to sulfa compounds, a third generation once-a-day dosing of oral cephalosporin or twice daily ciprofloxacin may be substituted, though there are no studies to provide specific guidance for these alternate prophylaxis agents.

When first started on itraconazole prophylaxis, liver function enzymes should be tested after the first few weeks to month of therapy. If levels are normal, then only yearly testing of liver enzymes is required. If liver enzymes are slightly to modestly elevated, the patient may be maintained on itraconazole, but should have liver enzymes tested every two weeks to follow the trend.

Most patients normalize their liver enzymes while on prolonged itraconazole and may be safely continued on this prophylaxis. Patients whose liver enzymes are very high, or continue to rise should be taken off itraconazole. A dose reduction re-challenge with itraconazole may be tried or the patient may not be able to take this agent over the long term.

Fluconazole lacks activity against Aspergillus species and may not substitute for itraconazole. The new high potency orally absorbed antifungals, voriconazole and posaconazole, have potent activity against Aspergillus and a range of molds that infect CGD patients. These agents have become a mainstay of treatment for systemic mold-related fungal infections in CGD patients, for the most part displacing even the liposomal amphotericin formulations, because of a relatively low toxicity profile.

Some have suggested that voriconazole or posaconazole replace itraconazole for fungus prophylaxis in CGD. However, long-term use of voriconazole can be associated with severe photosensitivity reactions in almost a third of patients, and posaconazole only comes in liquid formulation. Furthermore, there may be merit to retaining these potent agents for treatment rather than in use for prophylaxis. Nonetheless, in patients who have a documented history of several recurrent bouts of Aspergillus pneumonia, the use of voriconazole (or posaconazole) as prophylaxis in these selected patients may be warranted.

Recombinant human interferon gamma

The third component of infection prophylaxis that has been recommended for CGD is recombinant human interferon gamma (Actimmune®, currently distributed in the United States by InterMune, Inc., Brisbane, CA). An international double-blinded study involving 126 CGD patients participating from the United States and Europe evaluating three times weekly subcutaneous injections of interferon gamma, demonstrated that the time to subsequent infection was increased by about 70% in the interferon gamma group.

Despite this pivotal study, controversy remains about its use among some physicians, particularly in Europe, primarily because the medicine is very costly, requires injections, and has side effects. Furthermore, its mechanism of action in reducing infection has not been clearly determined. In practice, only about 25% of CGD patients in the United States choose to remain on long term prophylactic interferon gamma, either because of side-effects (fever, headache, malaise, difficulty in concentration at school or work, nightmares), copayment cost/availability, or discomfort with administration of injections.

Interferon gamma does not cure CGD and does not even improve the oxidase activity of neutrophils in most patients, but its action to decrease infection probably relates to a heightened surveillance function of the patient’s overall immune system. There are several single-case anecdotal reports of CGD patients not on prophylactic interferon gamma, whose severe infection improved after starting on gamma interferon as part of the treatment for an acute systemic bacterial or fungal infection.

However, there are no clinical trials to support or refute the use of interferon gamma as part of the treatment of an acute infection in a patient with CGD. Moreover, use of interferon gamma during major infection or major surgery can contribute to high fevers and hemodynamic instability, so some have counseled against administering interferon gamma in such settings.

More definitive therapies?

Bone marrow or stem-cell transplant

Allogeneic or matched unrelated donor bone marrow or other hematopoietic stem-cell transplant can cure CGD. There are an increasing number of CGD patients being referred for transplant in the United States and throughout the world, as methods of transplantation have improved to enhance engraftment, to better control and treat transplant associated infection, and to reduce the incidence and severity of graft versus host disease.

However, there is extreme variability in the frequency of infection, presence, or absence of inflammation complications of CGD, and the presence or progression of organ dysfunction in individual patients with CGD (even among affected siblings who have the same CGD mutation).

Prophylactic regimens

Furthermore, the outlook for survival and long periods of normal infection-free living for CGD patients continues to improve, such that many patients may experience several years to a decade between life-threatening infections. This has resulted from:

  • Effective prophylactic regimens

  • Aggressive approaches to infection diagnosis that include advance imaging technologies with imaging guided biopsies

  • Improvements in antibiotics

- Particularly the antifungal class, and early recognition of specific types of infections not previously known (such as Granulobacter bethesdensis)

  • Increasing appreciation for the salutary effect of low-dose alternate-day prednisone for managing granuloma/inflammation syndromes

New information to predict prognosis has helped in the decision-making for advising which patients should consider transplantation. Prognosis factors will be discussed in more detail below. However, in brief, any young child with CGD who has an oxidase function-null phenotype and who has a fully HLA-matched sibling donor should be considered for transplant.

Fully matched unrelated donor transplants or well-matched cord-blood transplants for CGD are still considered experimental, but molecular typing of HLA has allowed outcomes with these types of transplants to begin to approach those of matched sibling donors.

In older children and adults with a longer medical history to follow, bone marrow transplant should be a consideration for those who have frequent recurring life-threatening infections (more than one every 2 years). In addition to infection history and the level of residual oxidase activity measured in neutrophils, other prognostic factors correlated with decreased survival in older patients with observable multi-year medical history include the following:

  • Year to year decreases in platelet counts

  • One or more bouts of liver abscesses

  • Year to year decreases in pulmonary function, particularly decreasing room-air pulse-oxygen measurements

Studies of gene therapy for CGD have not, to date, resulted in cure or even long-term substantial improvements in CGD phenotype. However, this is an area of very active investigation that may benefit CGD patients at some time in the next 5 to 10 years. At this time, it should not be a consideration for selection of management options for CGD.

What other therapies are helpful for reducing complications?

A topic not previously covered in the other sections relates to the complications of granuloma formation and tendency for development of autoimmune disorders in CGD. Recognition of these problems in CGD patients and understanding management can be important in reducing complications from this aspect of CGD pathophysiology.

The inflammation/granuloma-related problems that affect CGD patients should be considered of three types:

  • Micro- and macro-granulomas not related to autoimmunity

Occasional non-caseating microgranulomas may be seen on histologic exam of biopsies in CGD patients of lung, lymph node, spleen, or colon that need not indicate infection or organ system pathology. Many young children may have chronic mild hepatomegally, mild splenomegally, and/or diffuse lymph node enlargement, that is not necessarily related to infection. However, granulomas with necrosis or caseation are almost always an indication of infection.

In some CGD patients, non-infection related macrogranulomatous inflammation in the stomach or pyloris may result in impaired emptying of the stomach or actual pyloric obstruction. Such dysfunction of gastric emptying particularly in children, may manifest as early satiety, occasional idiopathic vomiting, or simply as vague abdominal discomfort. Similarly, granuloma in the GU system may present, particularly in young boys, as acute onset of difficulty urinating or dull pain with urination.

Such partial or complete obstruction syndromes from non-infection granulomas of the GI or GU tract are very responsive to modest doses of 0.5mg/kg prednisone per day for a few days, followed by a taper over several weeks. If obstructive symptoms recur than patients may need to remain on 0.1-0.2mg/kg prednisone given every other day. Curiously, small alternate day doses of prednisone often is all that is needed as prophylaxis against recurrence of these CGD granuloma complications.

  • Crohn’s disease-like colitis

About 10 to 20% of CGD patients have modest to severe colitis, which can have all the features of Crohn’s disease, and should be managed together with a gastroenterologist, exactly as would be the standard of care for patients with Crohn’s disease. Patients can have rectal/sigmoid/colonic entero-cutaneous fistulas and colonic strictures as complications of this problem. Most patients require endoscopic evaluation and management with steroids and anti-tumor necrosis factor (TNF) biologics, plus other anti-inflammatory medications. However, anti-TNF agents carry significantly higher risk of infection when used in CGD patients.

  • Autoimmune disorders

CGD patients appear at higher risk than the general population of developing any of a number of defined autoimmune disorders that include, but are not limited to:

- Both cutaneous discoid and systemic forms of lupus erythematosis

- Sarcoidosis

- Idiopathic juvenile arthritis

- Rheumatoid arthritis

- Immune thrombocytopenia

- Anti-phospholipid hyperclotting syndrome

- IgA nephropathy- Immune thyroid diseases

- Many other autoimmune disorders. CGD patients who have signs, symptoms and laboratory values that satisfy the criteria for any of these autoimmune diseases should be treated with the established therapeutic modalities for the particular disorder.

What should you tell the patient and the family about prognosis?

Prompt diagnosis of CGD, diligent administration of prophylactic antibiotics, improved infection-diagnostic tools, and the discovery of both improved potent orally active antifungal agents, as well as improved antibacterial antibiotics, have greatly improved prognosis and patterns of mortality and morbidity of CGD.

Recent studies of large cohorts of CGD patients have demonstrated a very significant decrease in mortality in childhood so that the great majority of patients (greater than 90%) reach 20 years of age, and many are reaching late middle age. There is evidence that mortality rates tend to increase after age 20 in those patients with the most frequent infections, evidence of lung, or kidney impairment, and other poor prognostic factors. In this high-risk group, mortality may be in the range of 5% per year in the third decade of life and beyond.

As previously noted, overall survival appears to be predicted by the amount of residual oxidant production measured in neutrophils. However, other predictors of higher rates of mortality include history of liver abscess, year-by-year evidence of increasingly impaired liver, pulmonary or kidney function, and year-by-year decreases in platelet count. However, it must be noted that most CGD patients with the p47phox deficient autosomal recessive form of CGD and even the majority of those with X-linked CGD reach adulthood, go through college, get married and have families.

Thus, it is very important to keep CGD patients as healthy as possible to reduce the frequency of infections and inflammation complications. In addition, many CGD patients can have long infection-free periods of relative good health. Nevertheless, there are some patients who have frequent infections, and progressive organ impairment, despite good prophylaxis and good medical management. This makes it very difficult to decide which young children should be recommended for transplant or experimental high-risk therapies.

The decision of who should consider high-risk but curative options like transplantation becomes easier with the longer period of observation in older children and young adults. However, best transplant outcomes for CGD would be expected in younger children who are not infected at the time of transplant, and who have not yet accumulated organ damage from the infectious and inflammatory complications of CGD.

"What if" scenarios.

General guidance for infection diagnosis and management

Infections in CGD patients may vary greatly in their manifestations. Infections may or may not be associated with fever, neutrophilia, high erythrocyte sedimentation rate (ESR), or high C-reactive protein (though if they are elevated, following the levels may help to determine if treatment of the infection is effective).

Malaise may be the patient's only symptom, so it is important to pay attention to a patient or a patient’s parent who insists that the patient does not feel well, even where no fever or high white blood cell count is detected. Young children who seem active and happy with apparently normal physical examination, but where the parent insists something is just not right, often turn out to have a pneumonia on chest computerized axial tomograph (CAT) scan.

Fungal infection including pneumonia is often indolent and slow moving, may often present with only malaise and no fever or neutrophilia (or even as only an incidental finding of a small lung infiltrate on routine x-ray at routine annual or 6 month exam). High fever is more common in Nocardia, Serratia or Burkholderia bacterial species infections and may also be part of the presentation of inhalation related acute miliary fungal pneumonias, but may be absent early on in localized fungal infection.

Note that pneumonia in young CGD patients may present as abdominal pain, so children with CGD presenting with abdominal pain should also have a CAT scan of the chest in addition to any investigation of abdominal source. Early diagnosis using biopsy where possible to determine etiology should be a major goal of general management of CGD patients.

When etiology cannot be determined either because culture, gram stain, or biopsy histology was non-diagnostic, or because the infection was too small or in a difficult location for biopsy, then empiric oral antibiotics should be started. In general, levofloxacin or meropenem together with vancomycin or linezolid are good choices for empiric antibiotic anti-bacterial therapy in CGD. Voriconazole and/or an echinocandin are reasonable empiric anti-fungal treatment in CGD.

Some specific infection scenarios are noted and discussed below. In addition to the importance of determining the etiologic agent, a general principle of infection treatment in CGD is that prolonged antibiotic treatment for deep tissue infection is required for full resolution and to prevent relapse. As a rule, treatment for 3 weeks to 1 month for many infections is often required, and for fungal or nocardia pneumonias or for staphylococcal liver abscesses treatment for 3 to 6 months can sometimes be necessary; treatment can be guided by imaging.

  • Lymph node infection

Lymph node infections are common in CGD patients. Non-infectious lymphadenopathy is also common in CGD, but in that setting lymph nodes are not erythematous, tender, or fluctuant. Infection should be suspected if a lymph node shows erythema at the skin above the node, if the node is tender to palpation, or if fluctuance is noted. MRI imaging can be very helpful for detecting inflammation or abscess in the lymph node. In general, more rapid resolution and better outcome occurs with surgical removal of any lymph node that has drainable pus, rather than simply draining the node and putting in a wick.

  • Soft tissue infection

Most subcutaneous soft tissue infections with abscess in CGD result from Serratia or staphylococcal infections, but burkholderia, fungus and other pathogens may cause such infections.

  • Pneumonias

Pneumonias in CGD may be caused by a variety of bacterial or fungal pathogens that include but are not limited to burkholderia, klebsiella, nocardia, Serratia, aspergillus and demateaceous molds. Tissue diagnosis for histology and culture is important for CGD, because 50% of CGD pneumonias are fungal, and because 20% of pneumonias can be mixed infections (fungus and nocardia or burkhoderia species most common). At centers with good interventional radiology service, trans-thoracic needle biopsy under CT or fluoroscopy guidance yields an 80% diagnosis for fungal pathogen.

While bronchoscopy can often be diagnostic for bacterial infection in CGD, the yield for fungus infection in CGD is low and use of bronchoscopy alone may often miss the diagnosis of fungal infection. Even with biopsy, many fungi do not grow and are only seen on special fungal stain cytopathology exam of the needle biopsy or the special fungal stain histology of a core or larger biopsy sample.

Aspergillus fumigatis is a common fungal pathogen in CGD that is usually quite sensitive to prolonged treatment with voriconazole. However, increasingly, other more resistant species of Aspergillus and other relatively resistant species of fungi are being recognized as pathogens in CGD pneumonias.

Often these fungal pathogens such as some of the demateaceous molds including paecilomyces or geosmithia are found to be resistant to voriconazole and require treatment with posaconazole, with an echinocandin or combinations of these agents. Amphotericin, even the liposomal formulations of amphotericin, should be reserved for infections that are either not responding to an azole and/or an echinocandin antifungal, or where the organism is known to be resistant to these agents, because of concerns for long term effects on renal function.

It is useful to view fungal pneumonias as having two distinct patterns of presentation. The most common is the nodular/segmental fungal pneumonia in which one or more nodules occur or lung segments are involved that appear on x-ray or CT scan of the chest as a dense infiltrate with surrounding normal lung tissue. The usual presentation for such a pneumonia is as an indolent process in which malaise and fever appear slowly over several weeks. In some cases there may be no fever and just a cough or mild chest pain. Predisposing events are usually not determined.

The other type of fungal infection can be termed reticulo-miliary. Often there is a predisposing history in the previous 2 weeks of intense exposure to gardening mulch, digging in the soil, handling manure, or working on a construction project. The presentation is often that of relatively acute onset of shortness of breath and high fever. Early in the process a regular chest x-ray may appear normal, but a CT scan will often reveal pan lobular punctate miliary densities.

As the process progresses, these infiltrates will enlarge and coalesce. At the same time the patient may have a low blood oxygen saturation out of proportion to what might be expected from the extent of disease seen on the CAT scan. This type of presentation in a CGD patient should be considered a medical emergency requiring immediate attention. Since Aspergillus is often isolated from either bronchoscopy or transthoracic needle biopsy or even from sputum in this setting of inhalation pneumonia in CGD, this syndrome, as previously noted in an earlier section of this review, can be confused with that of the Aspergillus hypersensitivity pneumonitis (ABPA).

Like ABPA, part of the treatment must be with high dose steroids, which are necessary to achieve adequate oxygenation, which is impaired by the panlobular exuberant inflammatory granulomatous reaction associated with CGD. However, unlike ABPA, these patients also must be treated promptly and aggressively with anti-fungal agents such as intravenous voriconazole and/or an echinocandin to avoid high mortality. Inhalation pneumonias often have multiple organisms acting synergistically to cause pneumonia, and treatment should also include coverage for nocardia and burkholderia species.

  • Liver abscess

S. aureus at sites other than the liver or lymph nodes are no longer as common in CGD, yet almost 90% of liver abscesses in CGD patients are caused by S. aureus. The reason for this empiric observation is unclear, and it should not be assumed that a patient presenting with a staphylococcal liver abscess has been negligent about taking prophylactic trimethaprim-sulfamethoxazole.

In CGD patients, liver abscess is complicated by exuberant granuloma formation with a major fibrotic component. Thus, most liver abscesses are not a collection of pus surrounded by a capsule, but instead are a solid fibrotic mass studded with microabscesses and granulomas.

While needle biopsy can be used to make the microbiologic diagnosis, there is rarely more than a scant amount of pus drained from the site. Simple drainage of a CGD liver abscess is most often not a therapeutic option. While very prolonged antibiotic treatment (many months) can sterilize relatively small liver abscesses of this type, larger abscesses may require surgical extirpation of the fibrotic mass of micro-abscesses.

Recently it has been observed that with these staphylococcal liver abscesses, treatment with steroids (prednisone at about 0.5mg/kg) for a prolonged period (often 6 weeks or longer), together with antibiotics directed at the staphylococcus may result in resolution of the abscesses without need for surgery. It may be that the steroids prevent the fibrotic and granulomatous reaction from interfering with antibiotic penetration of the lesions and thereby enhance resolution.

  • Osteomyelitis

As previously noted, serratia osteomyelitis is seen particularly in infants with CGD. In older children and adults isolated osteomyelitis can also be seen, but is less common. Often osteomyelitis of the ribs or spine is the result of direct extension of a fungal pneumonia to the chest wall. In CGD patients with nocardia pneumonia or infection at other sites, it is important to do a brain scan and a bone scan, since cryptic metastatic infection with nocardia may occur at those sites.

  • Special pathogen – Granulobacter bethesdensis

In a CGD patient presenting with chronic fever, lymphadenopathy with or without splenomegally lasting for several weeks, who has a completely negative work-up to localize infection (that is, no pneumonia or specific localizing abscess), possible infection with Granulobacter bethesdensis should be considered. Sometimes positron emission tomography (PET) CAT scan can be used to find one or more lymph nodes that are strongly PET positive indicating infection.

Biopsy culture of the PET positive site, specifically culturing for this organism should be performed. Granulobacter is responsive to very prolonged treatment with ceftriaxone, though it can also respond to tetracyclines.

  • Comment on granulocyte transfusions for CGD

Granulocyte transfusions are only available where blood centers are set up to provide them. Once to twice weekly granulocyte transfusions can help in the management of an infection that has failed to improve, despite prolonged therapy with agents that should have been effective against the known pathogen. However, granulocyte transfusions are very expensive ($4,000 to $5,000 per transfusion). Also, because it is not possible to find HLA or neutrophil antigen matched donors, the development of limiting anti-HLA or anti-neutrophil antigen antibodies often occurs, eventually limiting the effectiveness of this treatment in patients.

For these reasons, granulocyte transfusions for treatment of CGD infection should be limited to those situations where conventional therapies appear to be failing. There are no controlled studies demonstrating efficacy of granulocyte transfusions to treat CGD infections, but there are many credible anecdotal reports of infections that failed to respond to maximum antimicrobial therapy that resolved in response to granulocyte transfusions.

Pathophysiology

The five genetic forms of CGD and the diagnostic tests to establish molecular type and mutation have been discussed in the sections above.

The primary biochemical defect in CGD is the failure to produce superoxide, hydrogen peroxide, hypochlorous acid, and other anti-microbial oxidants. Those infections particularly affecting CGD involve bacterial or fungal organisms for which phagocyte oxidant production is a critical factor in control of the infection.

It has long been asserted that organisms that produce high amounts of catalase (the enzyme that degrades hydrogen peroxide) cause the most infections in CGD patients. While there is some evidence to support this hypothesis, not all CGD pathogens fit this pattern, indicating that other virulence factors yet to be delineated are also important. Hydrogen peroxide also serves to activate some of the important proteolytic and other antimicrobial enzymes found in the granules of neutrophils.

In mouse genetic models of CGD, it has been shown that killed microorganisms or other non-pathogen initiators of inflammation, result in excessive ingress of neutrophils to those sites of inflammation, formation of granulomas, and failure of resolution of inflammation. This suggests that oxidant production by neutrophils is not only important for microbial killing, but that it also plays a role in the resolution of inflammation.

Some CGD patients fail to properly heal surgical incisions and such incisions may dehisce. Paradoxically, administration of low-dose steroids to CGD patients whose surgical wounds dehisce will help with healing. Thus, neutrophil oxidants may play a role in wound healing, and/or formation of granulomas in wounds may impair healing.

It is unclear why CGD patients have a higher incidence of some types of autoimmune disorders, but recent studies show that the phagocyte oxidase enzyme is present in very small amounts in T- and B-lymphocytes. This may play a role in the function of these lymphocytes. CD4+ T lymphocytes from CGD patients and from CGD mice have an increased tendency to acquire a Th1 or Th17 inflammatory phenotype, and it is possible that this may predispose CGD patients to higher penetrance of overt autoimmune disease driven by a genetic predisposition (that is, CGD acting as a potentiating cofactor upon a particular genetic background for predisposition to a particular autoimmune disorder).

There is an important genetic linkage of X-linked CGD to Mcleod blood phenotype. About 2 to 3% of patients with the X-linked genotype of CGD also have the McLeod blood phenotype (X-linked-Kx blood group). This is because the McLeod gene, XK, is located adjacent to the X-linked CGD gene, CYBB, and patients having both X-linked CGD and McLeod phenotype have a large deletion in the X-chromosome affecting both of these genes. This can be an issue for a patient with X-linked CGD who has some need for urgent blood transfusion where the Mcleod phenotype is not previously appreciated.

What other clinical manifestations may help me to diagnose chronic granulomatous disease?

It is important to obtain a family history. A history of early male deaths in the maternal lineage may indicate carrier status for X-linked CGD. A history of consanguinity or history of the mother and father coming from the same rural village may help in considering autosomal recessive forms of CGD. Recurrent pneumonias and/or lymph node infections, or osteomyelitis with Serratia or other unusual organism should trigger testing for CGD.

There are no pathognomonic findings on physical examination that specifically point to CGD. However, the differential diagnosis of any young child with significant splenomegaly or lymphadenopathy should include testing for CGD.

What other additional laboratory studies may be ordered?

Routine laboratory blood studies that should be performed every 6 months in CGD patients include complete blood count, including platelet count, renal and liver function studies, C-reactive protein and erythrocyte-sedimentation rate. Also useful is a room-air pulse-oximetry measurement while resting, and following a 6 minute walk.

What's the evidence?

Winkelstein, JA, Marino, MC, Johnston, RB. "Chronic granulomatous disease. Report on a national registry of 368 patients". Medicine. vol. 79. 2000. pp. 155-169.

[This report provides information that is still current, regarding the demographics of CGD and the problems that affect CGD.]

Kuhns, DB, Alvord, WG, Heller, T. "Residual NADPH oxidase and survival in chronic granulomatous disease". New Engl J Med. vol. 363. 2010. pp. 2600-10.

[This report provides current survival statistics for CGD and relates survival to the amount of residual oxidant production by neutrophils in CGD patients. It also reviews the types of mutations causing CGD and relates classes of mutations with residual oxidant production and survival.]

Jirapongsananuruk, O, Malech, HL, Kuhns, DB. "Diagnostic paradigm for evaluation of male patients with chronic granulomatous disease, based on the dihydrorhodamine 123 assay". J Allergy Clin Immunol. vol. 111. 2003. pp. 374-379.

[Reviews the utility of the DHR assay in diagnosis of CGD.]

Marciano, BE, Wesley, R, De, Carlo. "Long-term interferon-gamma therapy for patients with chronic granulomatous disease". Clin Infect Dis. vol. 39. 2004. pp. 692-699.

[This manuscript reviews the utility of interferon gamma as infection prophylaxis in CGD.]

Blumental, S, Mouy, R, Mahlaoui, N. "Invasive mold infections in chronic granulomatous disease: a 25-year retrospective survey". Clin Infect Dis. vol. 53. 2011. pp. 159-169.

[Excellent review of the European experience with fungal infections in CGD.]

Marciano, BE, Rosenzweig, SD, Kleiner, DE. "Gastrointestinal involvement in chronic granulomatous disease". Pediatrics. vol. 114. 2004. pp. 462-468.

[This manuscript reviews GI disease in CGD and includes management recommendations.]

Siddiqui, S, Anderson, VL, Hilligoss, DM. "Fulminant mulch pneumonitis: an emergency presentation of chronic granulomatous disease". Clin Infect Dis. vol. 45. 2007. pp. 673-681.

[This is a definitive description and discussion of management of inhalation fungal pneumonia in CGD.]

Leiding, JW, Freeman, AF, Marciano, BE. "Corticosteroid Therapy for Liver Abscess in Chronic Granulomatous Disease". Clin Infect Dis. vol. 54. 2011. pp. 694-700.

[This report reviews best current management recommendations for liver abscess in CGD.]

De, Ravin, Naumann, N, Cowen, EW. "Chronic granulomatous disease as a risk factor for autoimmune disease". J Allergy Clin Immun. vol. 122. 2008. pp. 1097-1103.

[This manuscript reviews the evidence for increased incidence of autoimmune diseases in patients with CGD.]

Seger, RA. "Hematopoietic stem cell transplantation for chronic granulomatous disease". Immunol Allergy Clin North Am. vol. 30. 2010. pp. 195-208.

[This is an excellent review of the current status of transplant for CGD.]
You must be a registered member of Clinical Advisor to post a comment.
close

Next Article in Hematology

Sign Up for Free e-newsletters