Download Whipple`s disease without malabsorption: new

Q J Med 1997; 90:765-772
Whipple's disease without malabsorption: new atypical
features
S.A. MISBAH, B. OZOLS\ A. FRANKS2 and N. MAPSTONE1
From the Departments of Chemical Pathology and Immunology, ^Molecular Pathology and
Neuropathology, Institute of Pathology, Leeds General Infirmary, Leeds, UK
2
Received 5 September 1997
Summary
The diagnosis of Whipple's disease in the absence
of intestinal involvement is difficult and often overlooked. We describe five patients aged 8-71 years
with normal jejunal biopsies and disparate clinical
features, previously unrecognized in Whipple's; all
were investigated at a single institution over a period
of 18 months. Routine histological examination for
periodic acid-Schiff (PAS) positive macrophages
and polymerase chain reaction (PCR) analysis for
Tropheryma whippelii was performed on the small
intestine in all patients. PCR analysis was also
performed on various tissues including peripheral
blood, lymph node, muscle, synovium and spleen in
individual patients. Patients 1, 2, 4 and 5 had
unusual presenting features not previously associ-
ated with Whipple's: intractable immune thrombocytopenic purpura (ITP), juvenile chronic arthritis,
isolated muscle weakness and quadriparesis,
respectively. Patient 3 presented with pyrexia of
unknown origin. All patients had histologically
normal small-bowel biopsies with no evidence of
PAS positive macrophages. PCR for T. whippelii was
positive in all patients in one or more tissues: peripheral blood, intestine, muscle, lymph node and
synovium. PAS-positive macrophages were found in
4/5 patients in various sites: lymph node, muscle,
spinal cord. Whipple's disease presents with protean
clinical features and should be considered in granulomatous disorders of unknown aetiology even in the
absence of gastrointestinal involvement.
Introduction
The diagnosis of Whipple's disease has traditionally
depended on the demonstration of diastase-resistant
periodic-acid-Schiff (PAS)-positive macrophages in
the small bowel of patients presenting with malabsorption and associated systemic features such as
fever, lymphadenopathy, arthritis, uveitis and neurological involvement.1 While rare cases of Whipple's
disease without obvious intestinal involvement were
recognized before the availability of the polymerase
chain reaction (PCR),2"6 this aspect of the disease
has received little attention. The recent identification
of the Whipple's bacillus, Tropheryma whippelii7'8
and the consequent development of diagnostic PCR
has led to the description of further cases of
Whipple's disease presenting without gastrointestinal
features. Recent case reports have drawn attention
to purely neurological and ocular cases, which in
the absence of PCR are likely to have been misdiagnosed.9'10 We describe a further five patients with
Whipple's disease presenting with disparate atypical
clinical features without gastrointestinal involvement
over a period of 18 months, that emphasize the
protean features of this disorder and the role of PCR
in diagnosis.
Case histories
Patient 1
A 22-year-old builder presented with fatigue,
hepatosplenomegaly, night sweats, lymphadenopathy (abdominal and supraclavicular), and fever in
Address correspondence to Dr S.A. Misbah, Department of Chemical Pathology and Immunology, Institute of Pathology,
Leeds General Infirmary, Great George Street, Leeds LSI 3EX
© Oxford University Press 1997
766
S.A. Misbah et al.
September 1991. A lymph-node biopsy revealed
multiple granulomata. In December 1991, he underwent laparotomy on account of continuing fever and
abdominal pain. Multiple granulomas were demonstrated in the para-aortic nodes. In February 1992, a
diagnostic splenectomy and further lymph-node
biopsies were performed. Multiple granulomas were
again noted in the spleen and lymph nodes. The
presence of a Gram-positive rod was noted on
splenic histology but cultures were sterile. Stains for
fungi, mycobacteria and PCR for M. tuberculosis
were negative. In April 1992, empirical treatment
with tetracycline was undertaken in view of continuing fever and sweats, on the premise of a bacterial
aetiology for his illness. For the duration of tetracycline therapy (6 months) he remained well, being able
to return to work. Marked thrombocytopenia (platelet
count 5 x 109/l) developed in May 1993, with resultant epistaxis. Bone-marrow biopsy revealed megakaryocytic hyperplasia with numerous granulomas.
Between May 1993 and June 1995, he had eight
episodes of profound thrombocytopenia requiring
high-dose corticosteroids. In June 1995, an immunological opinion was requested. A diagnosis of
Whipple's disease was considered at this stage, in
view of the history of fever, granulomatous lymphadenopathy, the presence of an unculturable Grampositive rod on splenic histology, and previous
clinical response to tetracycline therapy. PCR for T.
whippelii (details of assay described below) was
positive on peripheral blood, duodenum (despite no
histological abnormality) and sections of stored
spleen. Serology for cat-scratch disease was negative.
Sarcoidosis was considered unlikely on the basis of
a normal serum angiotensin-converting enzyme (104
U/l; NR 50-125) and a clear chest X-ray. Review of
his original lymph-node biopsies showed PAS-positive macrophages. Immunological investigations
revealed lgG-2 subclass deficiency associated with
sub-optimal responses to test immunization with 23
valent pneumococcal polysaccharide but normal
responses to tetanus and diphtheria toxoids.
Reduction of nitroblue tetrazolium (NBT) by the
patient's granulocytes was normal.
During a 9-month course of oral cotrimoxazole,
there was no obvious change in his well-being with
three further episodes of marked thrombocytopenia
and severe autoimmune haemolytic anaemia in
September 1995, necessitating the use of high-dose
steroids. He remained PCR-positive for T. whippelii
in peripheral blood and duodenum throughout this
period. Despite feeling well from January 1996,
cotrimoxazole was replaced by a 7-month course of
clarithromycin (500 mg b.d. for two weeks followed
by 250 mg b.d. maintenance therapy) in March 1996
in view of persistent PCR positivity. Four weeks after
clarithromycin treatment he was noted to be PCR-
negative on peripheral blood for the first time, and
able to undertake regular physical work. On review
in August 1997, he was free of systemic symptoms
but remains on low-dose steroids for ITP. His lgG-2
subclass deficiency worsened during the period June
1995 to July 1996 (0.77 to 0.23 g/l; reference range
1.2-6.6) but was not accompanied by bacterial
infection. Since January 1997, however, lgG-2 subclass levels have steadily improved to 1.01 g/l.
Patient 2
A 14-month-old baby girl presented in December
1987 with lymphadenopathy, hepatosplenomegaly,
rash and a raised ESR. A provisional diagnosis of
systemic Still's disease was made. Her subsequent
course was characterized by recurrent arthritis of her
knees and elbow joints associated with recurrent
upper respiratory tract infection and an episode of
orbital cellulitis in March 1991. In July 1992, she
underwent laparotomy for intestinal obstruction.
Numerous granulomas were noted in abdominal
lymph nodes, and her appendix was removed at that
time. Stains for acid-fast bacilli and fungi were
negative. Immunological investigations revealed a
polyclonally raised serum IgA but no evidence of
primary antibody deficiency, chronic granulomatous
disease or complement deficiency. In October 1995,
following the diagnosis in patient 1, the possibility
of Whipple's disease accounting for her unexplained
arthropathy, periodic fever and lymph node granulomas was considered. PCR on peripheral blood,
synovial and duodenal biopsies was positive for T.
whippelii. Both duodenal and synovial biopsies were
histologically normal. Review of the lymph-node
biopsy from July 1992 did not reveal PAS-positive
macrophages. Since commencing treatment with
cotrimoxazole, steady clinical improvement with a
decrease in the frequency and severity of flares was
noted, accompanied by PCR negativity on peripheral blood.
Patient 3
A 38-year-old man presented in May 1996 with a
four-week history of lethargy, fever, myalgia and
night sweats. An abdominal CT scan revealed marked
para-aortic lymphadenopathy and splenomegaly. An
abdominal lymph node biopsy obtained at diagnostic
laparotomy revealed massive sinus expansion with
large numbers of PAS positive macrophages. A jejunal
biopsy revealed normal villous architecture with no
evidence of PAS-positive macrophages, while a
muscle biopsy showed mild non-specific myopathic
changes. As with cases 1 and 2, PCR for T. whippelii
was positive on histologically normal intestine. PCR
analysis of peripheral blood and lymph node was
Atypical Whipple's disease
negative. Serology for brucellosis, mycoplasma,
human immunodeficiency virus infection, Lyme disease and toxoplasmosis was negative.
Treatment with oral cotrimoxazole was terminated
after a week due to the development of an allergic
rash. Tetracycline treatment led to rapid resolution
of symptoms enabling him to resume his hobby of
fell running.
Patient 4
A 72-year-old female presented in September 1995
with a 2-year history of painful, bilateral muscle
weakness associated with persistent sweating. On a
presumptive diagnosis of polymyalgia rheumatica,
steroid therapy had been started, but in view of
progressive muscle weakness and loss of mobility, a
muscle biopsy was performed. At the time she was
only able to walk a distance of 200 yards, due to
the pain in her legs. The muscle showed moderate
variation in fibre diameters, but no evidence of
interstitial inflammation, fibre necrosis or vasculitis.
Focal endomysial and perimysial infiltrates of macrophages were noted around areas of myotendinous
insertion (Figure 1); these contained granules that
were PAS-positive but negative with Gram, Giemsa,
Ziehl-Neilsen and desmin stains. PCR for T. whippelii
on fresh frozen muscle was positive. Electron microscopy of reprocessed paraffin sections of muscle
767
showed the macrophage granules to consist of
numerous aggregates of irregular fibrillar, dense material with occasional intact bacilliform structures.
The patient denied any gastrointestinal symptoms,
and no evidence of malabsorption was found on
investigation. A duodenal biopsy was normal, with
neither villous atrophy nor PAS positive macrophages. Oral cotrimoxazole was commenced,
leading to substantial clinical improvement and
resumption of normal daily activities. At follow-up 6
months later she had resumed competitive bowling.
In view of her steady clinical improvement, cotrimoxazole therapy was stopped at 14 months.
Patient 5
A 62-year-old, previously-well, female had a 7-week
history of neck pain with paraesthesiae in arms and
legs progressing to weakness due to a spastic quadriparesis with muscle strength of MRC grade 4 in the
upper limbs and grade 2 in both lower limbs. All
laboratory investigations, including CSF analysis were
normal; oligoclonal bands were not detected. An
MRI scan showed a high signal lesion in medulla
and cervical spinal cord, but no supratentorial
lesions. Empirical treatment with 12 mg dexamethasone daily was commenced but, despite an initial
response, her symptoms progressed, and after a
repeat MRI had shown extension of the cord lesion
Figure 1. Muscle biopsy in patient 4, showing numerous interstitial macrophages filled with PAS-positive granules. Periodic
Acid-Schiff stain, x 320.
768
S.A. Misbah et al.
from C3 to T2 level, a spinal-cord biopsy was
undertaken. This showed a widespread interstitial
infiltrate of foamy macrophages in which PAS- and
methenamine silver (Grocott)-positive bacilliform
structures were plentiful. A presumptive diagnosis of
Whipple's disease was made and a subsequent
jejunal biopsy, although morphologically normal,
was positive on PCR testing for T. whippelii.
Review of her history revealed no gastrointestinal
or systemic symptoms. She was treated initially with
a combination of intramuscular streptomycin and
oral cotrimoxazole. Cotrimoxazole was discontinued
on account of a drug eruption, and was replaced by
intravenous benzylpenicillin for 2 weeks followed
by maintenance oral penicillin. Her subsequent
course was complicated by a traumatic fracture of
the femoral neck necessitating arthroplasty. At her
latest follow-up, she was ambulant with a frame,
and able to feed herself while continuing to show
improvement in upper and lower limb function.
Table 1 summarizes the key clinical features in
these five patients.
Methods
Polymerase chain reaction
Fresh tissue specimens were subjected to standard
DNA extraction with proteinase K digestion and
phenol chloroform purification, followed by DNA
precipitation. Aliquots of 1 JLXI were then subjected
to PCR following the protocol of Relman et al.8
Specifically, the primers used were pW3FE and
pW2RB producing a 284 bp PCR product. The
resulting samples were then examined under ultraviolet light following ethidium bromide staining and
agarose gel electrophoresis, and a positive result
confirmed by the presence of a band of the correct
size (Figure 2).
All specimens were DNA extracted with negative
human DNA controls, which were subsequently
submitted for PCR at the same time as the sample
to exclude the possibility of contamination during
DNA extraction. PCR was performed with positive
controls and negative sample (water only) controls
to exclude the possibility of contamination during
setting up of PCR.
Discussion
We have described five patients with Whipple's
disease whose presenting features would not, in the
normal course of events, lead to a diagnosis of
Whipple's. The first three patients presented with
systemic disease characterized by lymphadenopathy,
fever and splenomegaly as common features. Both
pyrexia of unknown origin and granulomatous
lymphadenopathy are well-documented features of
Whipple's, but are not specific to this disorder.11'12
Relapsing thrombocytopenia was a prominent feature
in patient 1, and has previously been described in
only one patient with Whipple's. 13 In contrast,
marked thrombocytosis with platelet counts ranging
from 700 to 1100 x 109/l has been reported in some
patients.14 The mechanism of the thrombocytopenia
is unknown. If the Whipple's bacillus is present in
platelets, this might result in up-regulation of platelet
surface glycoproteins with consequent immune
complex formation and clearance by macrophages.
The cause of progressive lgG-2 subclass deficiency
in patient 1 is unclear. It is debatable whether this
is secondary to the disease process perse, or whether
it preceded the development of Whipple's. If the
latter, it could be argued that this patient exhibits
many of the features of common variable immunodeficiency (CVI).15 Classical Whipple's disease has
been documented in one patient with CVI, 16 while
agammaglobulinaemia was noted in the late stages
of the disease in another.17 These cases are exceptional, since serum immunoglobulin levels are reportedly elevated or normal in most patients.18 Although
various abnormalities of cellular immunity have been
described in Whipple's disease, including lymphopenia, increased expression of CD45 RO (marker of
memory T cells), decreased expression of CD11b,
impaired mitogen-induced lymphocyte transformation and reduced intracellular degradation of bacteria
by macrophages, it is unclear whether these abnormalities are primary or secondary to the disease
process.19'20 If a primary cellular immune defect is
present, it is likely to be subtle, since patients with
Whipple's disease are not prone to infection with
opportunistic pathogens, in contrast with other
cellular immune deficiency states.
The articular manifestations of Whipple's disease
have been well documented, commencing with
Whipple's original description in 1907.21 Joint
involvement is characterized in the main by a
migratory symmetrical non-erosive polyarthritis
affecting the knees, ankles and wrists.22 Where
synovial biopsies have been performed, a spectrum
of findings ranging from mild synovial cell hyperplasia to severe inflammation suggestive of septic arthritis has been reported.23'24 In contrast to the classical
rheumatological features, patient 2 had a long history
of ill health dating back to infancy, when a diagnosis
of juvenile chronic arthritis (JCA) was made. Whether
the onset of Whipple's disease in this patient coincided with her arthropathy in infancy is unclear.
Presentation of Whipple's disease in infancy is exceptionally rare, but has been reported in a three-monthold baby at autopsy.25 Although other multisystem
disorders may mimic JCA, there are no reports of
Episodic fever,
Hepatosplenomegaly,
granulomatous,
lymphadenopathy,
ITP, AIHA, IgG 2
subclass, deficiency
with impaired
response to
Pneumococcal
immunization
Arthropathy, episodic
fever,
hepatosplenomegaly,
granulomatous
lymphadenopathy
Swinging fever, night
sweats, splenomegaly,
para-aortic
lymphadenopathy
Proximal muscle
weakness, persistent
sweating
M
22
F
71
None
None
None
Single episode of
intestinal
obstruction at the
age of 6 yr
None
Gastro-intestinal
symptoms
No histological
abnormality
No histological
abnormality
PAS-positive
macrophages in
lymph node. Small
bowel negative
PAS-positve
macrophages in
muscle. Small
bowel negative
PAS-positive
macrophages in
spinal cord. Small
bowel negative
ND
Pos
ND
ND
ND
ND
ND
Muscle
Neg
Pos"
Negative in
synovium lymph
node and small
bowel
No histological
abnormality
No histological
abnormality
Pos"
Blood
PCR
PAS-positive
macrophages in
lymph node. Small
bowel negative
PAS staining
of tissues
No histological
abnormality
Jejunal biopsy
Pos
ND
Pos
Pos
Pos
Intestine
* Reverted to negative following treatment. ITP, immune thrombocytopenic purpura; AIHA, autoimmune haemolytic anaemia; ND, not done.
62
M
38
Spastic quadriparesis
Clinical features
Sex
Age
(years)
Patient
F
Summary of key clinical features
Table 1
ND
ND
ND
Pos
ND
ND
ND
ND
ND
Pos
Synovium Spleen
ND
ND
Neg
ND
ND
Lymph
node
770
S.A. Misbah et al.
Figure 2. Agarose gel electrophoresis of PCR-amplified
DNA. Lanes 1 and 10, a 100 bp sized DNA ladder. Lane
2, positive control from a case of known gastrointestinal
Whipple's showing a 284 bp PCR product. Lane 6, an
identical PCR product from a representative patient in the
study. Lanes 4, 8 and 9, negative controls comprising
water (lane 4), human DNA extracted at a different time
(lane 8) and human DNA extracted at the same time
(lane 9) as the patient sample.
Whipple's disease being mistaken for JCA. Despite
histologically normal jejunal and synovial biopsies,
patient 2 was PCR-positive in these tissues as well
as in peripheral blood.
The clinical presentation of patient 4 with isolated
muscle weakness in the absence of systemic features
suggestive of Whipple's is highly unusual, and has
not been described before. Despite the absence of
significant histological abnormality of the muscle
fibres, focal endo- and perimysial infiltrates of PASpositive macrophages were observed, coupled with
a positive PCR signal in muscle for T. whippelii. It
is tempting to attribute the myalgic symptoms to the
localization of inflammatory cells at the myotendinous junction where golgi tendon organs are present.
Clinically overt myopathy has only been described
previously in one patient with Whipple's disease.
Swash et al. reported a 51-year-old man with antibiotic-responsive proximal myopathy associated with
malabsorption and migratory arthralgia, in whom
muscle biopsy showed mild type 2 fibre atrophy and
PAS-positive macrophages.26 Tubular aggregates of
uncertain nature in normal muscle without the
presence of PAS-positive macrophages have also
been described in a single patient with classical
Whipple's. 27 The specificity of this finding for
Whipple's disease is unclear, given the wide range
of disorders in which tubular aggregates in muscle
have been described.28 The appearances of interstitial
macrophages with PAS-positive granules are similar
to those illustrated by Banker and Engel, although
clinical details were not given in this case.21
The neurological complications of Whipple's disease have been well described, with over a third of
patients presenting with the characteristic triad
of dementia, external ophthalmoplegia and facial
myoclonus.29 Neurological involvement may either
form part of systemic disease or present de novo
without apparent gastrointestinal involvement. While
such a presentation with the characteristic neurological triad would lead neurologists to consider the
diagnosis of Whipple's, pure spinal involvement as
in patient 5 would not. Despite the complete absence
of gastrointestinal symptoms and a morphologically
normal small bowel, PCR positivity for T. whippelii
was demonstrated on jejunal biopsy.
The use of PCR to demonstrate the presence of T.
whippelii in peripheral blood mononuclear cells in
addition to lymph nodes, duodenum and vitreous
humour has raised the possibility of it being used as
a non-invasive diagnostic test for Whipple's disease.
PCR positivity in histologically-normal small bowel
was noted in four of our patients and has also been
recorded by others.10 This is an important observation
which suggests the presence of sub-clinical bowel
involvement in some patients. In patients who are
PCR-positive on peripheral blood at initial assessment, it may prove to be a useful means for
monitoring response to treatment. Enthusiasm for the
use of PCR in the diagnosis and monitoring of
Whipple's disease does not, however, mean that the
search for PAS-positive macrophages can be discarded since Marth et al. have recently reported
negative PCR results in peripheral blood from eight
patients with Whipple's disease30 (4 with active
disease, 4 with treated inactive disease).
Whether PCR analysis should supplant the demonstration of PAS-positive macrophages as the gold
standard for the diagnosis of Whipple's disease is
unclear at present, and requires prospective study.
We failed to demonstrate PAS positivity in one of
our patients (patient 2) despite positive PCR signals
from peripheral blood, duodenum and synovium.
Similarly, Cohen et al. reported a case of purely
neurological Whipple's diagnosed on the basis of
PCR positivity on cerebrospinal fluid with no evidence of PAS-positive cells either in the CSF or small
intestine.9 Given the atypical clinical presentations
in these cases we did consider the possibility that
our PCR results were artefactual. However, we
believe that the unequivocal presence of PAS-positive
macrophages in four patients, reversion of PCR
positivity with treatment in 2/5 patients and the
clinical response to prolonged antibiotic treatment
suggests that this was unlikely to be the case.
The five patients that we have described illustrate
the wide heterogeneity of clinical presentation of
Atypical Whipple's disease
Whipple's disease and the considerable difficulties
associated with making a diagnosis when the clinical
presentation is atypical. Even when Whipple's disease is considered as part of the differential diagnosis,
the absence of gastrointestinal involvement clinically,
as in our patients, and negative PAS staining of small
bowel may lead to the diagnosis being rejected. We
suggest that the possibility of Whipple's be considered in any patient with an unexplained granulomatous disorder, even in the absence of gastrointestinal
involvement. Our experience also suggests that the
quest for PAS-positive macrophages
in suspected
cases should be accompanied by PCR analysis of
relevant tissue.
The pattern of clinical, pathological and laboratory
findings in these cases suggests that we should view
the relationship between man and the Whipple's
bacillus as a spectrum. This ranges from the presence
of the bacillus with no discernible
morphological
reaction (but detectable with PCR), through a localized inflammatory
reaction at the site of
initial
infection (the classical intestinal form) to a more
widespread dissemination of the bacillus (detectable
by PCR) but with variable reactions in a range of
tissues, which may include a necrotizing macrophage
response or a granulomatous response. What it is
that determines the pattern in an individual is conjectural, but immune competence and infective doses
will be important variables. We do not know how
widespread exposure to the bacillus which causes
Whipple's disease is, or how many cases of infection
occur without clinical disease. The availability of a
sensitive tool in the form of PCR opens up the
possibility
of
conducting
the
population
studies
necessary to answer these questions.
We are extremely grateful to Dr M. Galvin for
patient
1,
Drs
with negative jejunal histology. N Engl J Med 1979;
300:907-8.
4. Adams M, Rhyner PA, Day J, ef al. Whipple's disease
confined to the central nervous system. Ann Neurol 1987;
21:104-8.
5. Wroe SJ, Pires M, Harding B, et al. Whipple's disease
confined to the CNS presenting with multiple intracerebral
mass lesions. J Neurol Neurosurg Psychiatry 1991;
54:989-92.
6. Adler CH, Caletta SL. Oculo-facial-skeletal myorhythmia in
Whipple's disease: treatment with Ceftriaxone. Ann Intern
Med 1990; 112:467-9.
7. Wilson KH, Blitchington R, Frothingham R, Wilson JAP.
Phylogeny of the Whipple's disease associated bacterium.
Lancet 1991; 338:474-5.
8. Relman DA, Schmidt TM, MacDermott RP, Falkow S.
Identification of the uncultured bacillus of Whipple's
disease. N Engl J Med 1992; 327:293-301.
9. Cohen L, Berthet K, Dauga C, ef al. Polymerase chain
reaction of cerebrospinal fluid to diagnose Whipple's
disease. Lancet 1996; 347:329.
10. Rickman LS, Freeman WR, Green WR, ef al. Uveitis caused
by Tropheryma whippelii (Whipple's bacillus). N Engl J Med
1995; 332:362-6.
11. Fleming JL, Wiesner RH, Shorter RC. Whipple's disease:
Clinical, biochemical and histopathologic features and
assessment of treatment in 29 patients. Mayo Clin Proc
1988; 63:539-51.
12. James DG, Williams WJ. Report of the Commission on
granulomatous diseases. Ann NY Acad Sci 1986;
465:304-12.
13. Finelli PF, McEntee WJ, Lessell S, ef al. Whipple's disease
with predominantly neuroophthalmic manifestations. Ann
Neuron 977; 1:247-52.
14. Nuzum CT, Sandier RS, Paulk HT. Thrombocytosis in
Whipple's disease. Gastroenterology 1981; 80:1465-7.
15. Spickett GP, Misbah SA, Chapel HM. Primary antibody
deficiency in adults. Lancet 1991; 337:281-4.
16. Cochran M, Cook MG, Gallagher JC, Peacock M.
Hypogammaglobulinaemia with Whipple's disease.
Postgrad Med} 1973; 49:355-8.
Acknowledgements
referring
771
S. Wyatt,
D.
Spence,
C.J. Mitchell, C.E. Clarke and C.E. Millson for permission to report on their patients, Drs Andrew Jack and
M. Dixon for histological reports, Dr S. Chitturi for
serial endoscopies on patient 1, Dr D. Veale for
synovial biopsy on patient 2, and to Faye Storey for
typing the manuscript.
References
1. Dobbins WO III. The diagnosis of Whipple's disease. N Engl
JMed 1995; 332:390-2.
2. Mansbach CM, Shelburne JD, Stevens RD, Dobbins WO III.
Lymph node bacilliform bodies resembling those of
Whipple's disease in a patient without intestinal
involvement. Ann Intern Med 1978; 89:64-6.
3. Feurle CE, Volk B, Waldherr R. Cerebral Whipple's disease
17. Anton AT. Agammaglobulinaemia complicating Whipple's
disease. Ohio State Med} 1961; June:650-2.
18. Dobbins WO III. Is there an immune deficit in Whipple's
disease? Dig Dis So 1981; 26:247-52.
19. Marth T, Roux M, von Herbay A, ef al. Persistent reduction
of complement receptor 3 *-chain expressing mononuclear
blood cells and transient inhibitory serum factors in
Whipple's disease. Clin Immunol Immunopathoh 994;
72:217-26.
20. Bjerkness R, Odergaard S, Bjerkvig R, ef al. Whipple's
disease—demonstration of a persisting monocyte and
macrophage dysfunction. Scand} Gastroenterol 1988;
23:611-19.
21. Whipple GH. A hitherto undescribed disease characterised
anatomically by deposits of fat and fatty acids in the
intestinal mesenteric lymphatic tissues. John Hopkins Hosp
Bull 1907; 18:382-91.
22. Weiner SR, Utsinger PD. Whipple's disease. Seminars in
Arthritis and Rheumatism 1986; 15:157-70.
23. Caughey DE, Bywaters EG. The arthritis of Whipple's
syndrome. Ann Rheum Dis 1963; 22:327-35.
772
S.A. Misbah et al.
24. Rubinow A, Canoso JJ, Goldenberg DL, Cohen AS. Arthritis
in Whipple's disease. IsrJ Med Sci 1981; 17:445-50.
25. Aust CH, Smith EB. Whipple's disease in a three month old
infant. AmJClin Pathol 1962; 37:66-74.
26. Swash M, Schwartz MS, Vandenburg MJ, Pollock
DJ. Myopathy in Whipple's disease. Cut 1977; 18:800-4.
27. Stamboulis E, Manta P, Kararizou E, Crivas L. Whipple's
disease with tubular aggregates in asymptomatic muscle.
Clin Neuropathoh 993; 12:121-4.
28. Banker BQ, Engel AC. Basic reactions of muscle. In Engel
AC, Franzini-Armstrong C, eds. Myology, 2nd edn. New
York, McCraw Hill, 1994:832-88.
29. Dobbins WO III. Whipple's disease. Springfield, Charles C
Thomas, 1987:113.
30. Marth T, Fredricks D, Strober W, Relman DA. Limited role
for PCR-based diagnosis of Whipple's disease from
peripheral blood mononuclear cells. Lancet 1996;
348:66-7.