The management of patients with a chronic multisystem infection is more complex. Once a diagnosis of intestinal Whipples disease is reached, it is important to evaluate other commonly involved organ systems. After staging is completed, the choice and adequate duration of antibiotic treatment is critical for the patient’s long term outcome. Careful monitoring is needed to ascertain its efficiacy.
Diagnostic studies should evaluate tissues along the presumed route of infection, which starts with bacterial invasion in the upper small intestine (Table Proposal for Staging of Whipple’s Disease). Beyond endoscopy, the draining mesenteric nodes and other retroperitoneal lymph nodes should be investigated by abdominal sonography. Enlarged nodes with lipid deposits (which result in high echogenicity) are a special sign of intestinal Whipples disease. Physical examination should focus on the presence of peripheral lymphadenopathy on evidence of cardiac involvement (pericardial effusion, valvular murmur), and also look for skin hyperpigmentation and evaluate body weight and temperature. Chest radiography should be performed, looking at heart diameter, possible pericardial calcifications, and mediastinal lymphadenopathy. Echocardiography is generally not helpful in patients without a murmur and normal chest radiography.
TABLE. Proposal for Staging of Whipple’s Disease
|Stage||Location of Documented Manifestation|
|lb||Abdomincal lymph nodes|
|II||Extraabdominal lymph nodes|
|III||Extraabdominal orans (brain, heart, others)|
An essential part of staging is cerebrospinal fluid analysis, even in the absence of neurological or psychiatric symptoms. One of the new insights gained during the 1990s was the recognition that, at the time of first diagnosis, approximately 70% of untreated patients with intestinal Whipples disease have asymptomatic (“silent”) central nervous system infection with T. whippelii. These patients are considered to be at high risk of progressing from silent central nervous system infection to symptomatic central nervous system Whipples disease (or cerebral Whipples disease) unless the bacteria are eradicated. cerebrospinal fluid analysis should be performed in parallel by means of polymerase chain reaction (PCR), which detects bacterial DNA, and by cytology of a large cerebrospinal fluid sample which detects highly characteristic sickle-form particle containing cells. In practice, we recommend obtaining a 10 mL sample of cerebrospinal fluid where possible; 8 mL of fresh cerebrospinal fluid should be immediately prepared by cell concentration techniques for cytological study with PAS stain, while an aliquot of 2 mL should be fresh frozen and forwarded for PCR analysis. In our personal experience, the diagnostic yield of cerebrospinal fluid cytology is roughly similiar with PCR analysis, but it is much less effective with samples < 5 mL. Imaging studies of the brain (computed tomography, magnetic resonance imaging) are generally not helpful in Whipples disease patients without neurological symptoms.
Additional Diagnostic Studies
Peripheral blood analysis does not contribute to staging, but it can provide other information. Most patients present with laboratory signs of inflammation (eg, elevated erythrocyte sedimentation rate, C-reactive protein), many have laboratory signs of malabsorption (eg, low iron, lowbetacarotene), and some may have occult intestinal bleeding (eg, low hemoglobin, low iron). Routine laboratory tests of liver function (ie, aspartate aminotransferase, alanine aminotransferase, y-GT, AP, and albumin), renal function (ie, creatinine, sodium, potassium, and calcium), and bone marrow function (ie, blood cell counts) should be performed to recognize possible problems with drug availability.
Although an immunological deficit is likely to play a role in the pathogenesis of Whipples disease, diagnostic studies of immunological functions in individual patients are of uncertain value. A variety of cellular and humoral abnormalities may be observed. Serum electrophoresis may recognize hypogammaglobulinemia or hypergammaglobulinemia.
Response of intestinal Whipples disease can be achieved with many drugs (Table Overview of Antibiotics Used for the Treatment of Whipple’s Disease), but the real challenge is to prevent late onset cerebral Whipples disease. To reach this goal, our first line recommendation is to combine an induction therapy with intravenous application of a bactericidal drug, ceftriaxone (2 times 1 mg/d) for 14 days, and a maintenance treatment with trimethoprim-sulphamethoxazole (2 times 160 mg, 2 times 800 mg/d by mouth) for 12 months.
Ceftriaxone, a third-generation cephalosporin, is commonly used to treat bacterial meningitis. We have 10 years of experience with its use in Whipples disease.
TABLE. Overview of Antibiotics Used for the Treatment of Whipple’s Disease
|Ceftriaxone||3 mg/d IV||Induction therapy (first 14 days), good central nervous system penetration|
|Meropeneme||3 mg/d IV||Induction therapy (first 14 days), good central nervous system penetration, limited experience|
|Penicillin G plus streptomycin||6 to 24 million units/d IV plus 1 g/d IM||nduction therapy (first 7 to 14 days), moderate central nervous system penetration|
320 mg/d P0,
1600 mg/d P0
|Maintenance therapy, first line drug, good central nervous system penetration, but central nervous system Whipples disease may occur|
|Doxycycline (ortetracycline)||100 to 200 mg/d P0||Maintenance therapy; divergent reports of low and high rate of clinical relapses|
|Penicillin VK||500 mg 4 times daily P0||Alternative for maintenance therapy, limited experience|
|Cefixime||400 mg twice daily P0||Alternative for maintenance therapy, limited experience|
|Rifampin||BOO mg 4 times daily P0||Second line drug, good central nervous system penetration, limited experience in central nervous system Whipples disease|
|Chloramphenicol||1000 mg 4 times daily P0||Second line drug, worrisome side effects|
|Erythromycin||500 mg 4 times daily P0||Second line drug, limited experience|
Alternatively, meropeneme (1 mg 3 times a day) may be given, but our experience with this drug in Whipples disease is limited to a 4-year experience. The historical “Duke regimen”, previously reported to be effective, might be considered as a further alternative, although penicillin and streptomycin are considered to only moderately penetrate the normal blood-brain barrier.
Apart from common practice, our argument in favour of 1 year of maintenance treatment is derived from follow-up studies of patients with PCR analysis. Clearance of bacterial DNA from the intestinal mucosa occurs within a range of 1 to 12 months, but, in our experience, after 1 year stable intestinal remission is virtually always reached. An other argument for prolonged therapy can be derived from experiments to culture T. whippelii, where the growth of the bacteria is very slow. The estimated duplication time of 4 days is fairly different from common enteric bacteria (eg, Escherichia coli has a doubling time of approximately 20 minutes). Surprisingly, there are a few reports of short term antibiotic treatment resulting in long lasting remissions. This paradox is puzzling.
During antibiotic treatment, remission of symptoms and of abnormal diagnostic findings should be followed and documented at regular intervals. The initial response of patients is usually prompt. Diarrhea often resolves within several days, arthralgias within a few weeks, and significant weight gain occurs within a few months. When this clinical improvement is occuring, noninvasive laboratory examinations may be sufficient during the first 6 months.
A potential pitfall of monitoring patients is that intestinal biopsies do not equate with extraintestinal Whipples disease. It requires multiple invasive follow-up examinations in parallel. In brief, monitoring repeats staging examinations.
After 6 months, cerebrospinal fluid analysis should be repeated, either to confirm eradication of previously detected silent central nervous system infection with T. whippelii or to ascertain its absence. Although the bactericidal effects of antibiotics will arise much earlier, there is a lag of clearance of bacterial DNA from the cerebrospinal fluid. Up to several weeks, DNA fragments may still be detected by PCR. During this lag phase, interpretation of positive PCR results is difficult and may result in the erroneous impression of treatment failure. Similiarly, interpretation of cytologic findings is difficult during treatment and should be jointly performed with PCR results. Endoscopy with biopsies should be repeated. Although grossly visisble mucosal lesions may or may not be absent, histology of biopsies should document partial remission of mucosal lesions, which indicates the common response to treatment. At this time intestinal mucosal remission is always incomplete, but subtypes of PAS positive macrophages have significantly changed. Abdominal sonography should expect to indicate partial, but still incomplete, remission of abdominal lymphadenopathy
After 12 months, before stopping treatment, cerebrospinal fluid analysis should be repeated to ascertain eradication of previously detected silent central nervous system infection with T. whippelii. This is of critical importance for the patient’s outcome. At this time, a DNA-negative result from PCR analysis is more important than the cytologic detection of PAS positive cells, because subtyping of PAS-positive macrophages in the cerebrospinal fluid is not established. Endoscopy will usually be normal. When multiple biopsies are obtained, histology will document intestinal remission, although a variable number of PAS positive macrophages may still be present. Subtyping of PAS macrophages can reliably differentiate between remission and persisting infection. This can be confirmed by subsequent PCR analysis. Abdominal sonography may document complete remission of previous lesions, but, in some patients, remission is still incomplete and fibrosis may remain.
After remission is established, the physician and patients should be alert to the possible recurrance of symptoms. If this happens, it does not necessarily indicate a relapse of Whipples disease. A tissue-based diagnosis of relapse should be attempted.
Occasionally, diarrhea may persist (or recur) despite some weeks of treatment. We have found that this is suggestive of an associated infection with Giardia lamblia, rather than necessarily being refractory (or relapsing) Whipples disease. Associated G. lamblia may easily escape notice in PAS stained biopsies. This protozoan requires other drugs to become eradicated.
Rarely, a patient with intestinal Whipples disease does not respond to adequate primary treatment and does not reach clinical and mucosal remission after 1 year. In our personal series, there are 2 nonresponders among 162 patients. Salvage therapy should be planed on an individual basis.
As mentioned, the real challenge of treatment is to eradicate asymptomatic central nervous system infection. In those patients who do not respond to initial treatment with ceftriaxone, a second course with meropeneme should follow. Further salvage therapy should be planned on an individual basis.