MedicineLakex

 

Cuso.ch

Tropheryma whipplei Bacteremia during Feverin Rural West Africa Florence Fenollar,1 Oleg Mediannikov,1,2 Cristina Socolovschi,1 Hubert Bassene,2 Georges Diatta,2 Herve´ Richet,1
Adama Tall,3 Cheikh Sokhna,2 Jean-Franc¸ois Trape,2 and Didier Raoult1,2
Unite´ de Recherche sur les Maladies Infectieuses et Tropicales Emergentes (URMITE), Universite´ de la Me´diterrane´e, Faculte´ de Me´decine,1Marseille, France, and 2Dakar, Se´ne´gal; and 3Institut Pasteur de Dakar, Dakar, Se´ne´gal (See the editorial commentary by Greub, on pages 522–524.)
Tropheryma whipplei not only causes Whipple disease but also is an emerging pathogen associated with gastroenteritis and pneumonia that is commonly detected in stool samples in rural West Africa. We investigatedthe role of T. whipplei in febrile patients from rural Senegal who had a negative test result for malaria.
From November 2008 through July 2009, we conducted a prospective study in 2 Senegalese villages; 204 blood specimens from febrile patients were collected. DNA extraction of whole-blood samples collected byfinger pricks with a lancet stick was performed in Senegal; elution and quantitative polymerase chain reactionassays for T. whipplei were performed in France. In April 2009, we conducted a screening to look for the presenceof T. whipplei in the saliva and stools of the overall population. Blood from French patients with chronic T. whippleiin stool samples was also analyzed.
The presence of T. whipplei DNA was detected in blood from 13 (6.4%) of 204 tested patients, mostly in children and in December and January. None of the French carriers tested positive. The patients with T. whippleibacteremia presented with fever (13 patients), cough (10), thirst (8), fatigue (7), rhinorrhea (6), and sleep disorders (5). Cough and sleep disorders were significantly more frequent in febrile carriers than in the 191 febrile episodeswithout T. whipplei bacteremia (P p .002 and .005, respectively). No correlation was observed between the presenceof T. whipplei in the stools and saliva and bacteremia.
Our findings suggest that T. whipplei is an agent of unexplained cold season fever with cough in rural West Africa.
Until recently, Tropheryma whipplei was thought to be itis, and pneumonia [1, 3, 4, 6–8]. The bacterium is a rare bacterium typically causing the classic form of also highly prevalent (15%) in fecal samples obtained Whipple disease, which is characterized by histologic from children aged 2–4 years in France who have gas- periodic acid–Schiff–stained bacilli in infected small- troenteritis but was not detected in samples from a bowel macrophages [1, 2]. However, this well-known control group of children of the same age without di- entity, described mainly in white men, represents only arrhea [9]. T. whipplei DNA has also been detected in 1 rare clinical manifestation of T. whipplei infection.
stool and saliva specimens from asymptomatic individ- Indeed, recent studies have shown that the spectrum uals, but its prevalence depends on the geographic area of infections due to T. whipplei is wide [1, 3–5] because [10, 11]. In Europe, the prevalence of this bacterium this bacterium can cause localized infections, such as in fecal samples from the healthy general adult popu- endocarditis, spondylodiskitis, meningoencephalitis, uve- lation is estimated to be 1%–11% [1]. In addition, thebacterium has been detected in sewage and is moreprevalent in fecal samples of sewer workers (12%–26%) Received 17 March 2010; accepted 29 April 2010; electronically published 26 than in the general population [12–14]. In a recent Reprints or correspondence: Prof Didier Raoult, Unite´ de Recherche en Mal- study, 44% of children aged 2–10 years who live in adies Infectieuses et Tropicales Emergentes (URMITE), UMR CNRS-IRD 6236–198, rural Senegal carried T. whipplei in their stool samples Faculte´ de Me´decine, 27 Bd Jean Moulin, 13385 Marseille Cedex 5, France([email protected]).
[15]. Water from all 8 village wells have tested negative Clinical Infectious Diseases
for T. whipplei by polymerase chain reaction (PCR)  2010 by the Infectious Diseases Society of America. All rights reserved.
[15]. This research was part of a longitudinal study of the host-parasite relationship in 2 Senegalese villages, T. whipplei Bacteremia • CID 2010:51 (1 September) • 515
which was initiated in 1990 and focused initially on malaria washing with Buffer AW2 and incubation at room tempera- and tickborne borreliosis [16, 17]. Improvements in the knowl- ture for 5 min with 100 mL of Buffer AE, DNA was eluted by edge and management of these diseases have led us to extend centrifugation at 6000g for 1 min.
our studies to look for other etiologic agents of fever. Because Master mixtures were prepared by following the manufac- T. whipplei is highly prevalent in human stool samples obtained turer's instructions. Quantitative real-time PCR was performed in rural Senegal and is also an agent of gastroenteritis [9], we using a LightCycler instrument (Roche Diagnostics) with the first hypothesized that T. whipplei gastroenteritis may corre- QuantiTect Probe PCR Kit as described by the manufacturer.
spond to a primo-infection and that there was bacteremia in First, specimens were tested using the Twhi3F: 5
-TTGTGTA- patients who presented with fever [9]. Thus, our aim was to TTTGGTATTAGATGAAACAG-3
and Twhi3R: 5
-CCCTACA- investigate the potential of T. whipplei to cause infection in ATATGAAACAGCCTTTG-3
primer pair and the specific patients with unknown fever and a negative test result for ma- TaqMan probe Twhi3: 6-FAM-GGGATAGAGCAGGAGGTGT- laria in rural Senegal.
CTGTCTGG-TAMRA, as previously reported [8]. When thespecimen tested positive in this assay, the result was confirmed PATIENTS, MATERIALS, AND METHODS
by a second quantitative PCR using the Twhi2F: 5
-TGAGGAT- GTATCTGTGTATGGGACA-3
and Twhi2R: 5
-TCCTGTTAC- In 2008 and 2009, we performed AAGCAGTACAAAACAAA-3
primer set and the Twhi2 probe: studies among the population of Dielmo (1343
N, 1625
W) 6-FAM-GAGAGATGGGGTGCAGGACAGGG-TAMRA. A case and Ndiop (1433
N, 1615
W), 2 villages in the Sine-Saloum was defined as 2 positive quantitative PCR results in assays region of Senegal, to assess the incidence of T. whipplei. These targeting 2 different T. whipplei repeated DNA sequences. The villages are included in the Dielmo project, a longitudinal pro- quality of DNA handling and extraction of blood samples was spective study initiated in 1990 for long-term investigation of verified by quantitative PCR for a housekeeping gene coding host-parasite associations [16, 17]. In 2009, the survey showed for b-actin, as previously described [18].
a population of 391 in Dielmo (ratio of female to male subjects, Stool and saliva analysis.
In April 2009 we looked for T. 1.04:1), of which 35% were children younger than 10 years,and a population of 313 in Ndiop (ratio of female to male whipplei in stool and saliva of the overall population, as pre- subjects, 1.1:1), of which 36.5% were children aged viously done and reported in April 2008 [16]. We also sampled At the beginning of the study, all members of the population, stool samples from patients when they developed diarrhea. Af- including patients and parents or legal guardians of all children, ter collection, each fecal specimen was mixed with 2.5 mL of gave written individual informed consent. The national ethics absolute ethanol and each saliva specimen with 1 mL of absolute committee of Senegal and the local ethic committee of the ethanol for storage and transportation to our laboratory at Institut Fe´de´ratif de Recherche 48 (Marseille, France) approved room temperature. On arrival, DNA was extracted using the this study. Detection of individuals with fever was both passive BioRobot MDx workstation (Qiagen) in accordance with the and active. The dispensary created for the project was open 7 manufacturer's recommendations and protocols. T. whipplei days a week. Every villager was also visited daily at home for quantitative PCR assays were performed as described herein.
clinical monitoring. For individuals with fever (defined as an A case was defined as 2 positive quantitative PCR results in axillary temperature 137.5C), a medical attendant completed assays targeting 2 different T. whipplei DNA sequences.
a medical examination and a thick blood film.
Blood control group.
Ten blood specimens received in our Whole-blood samples were collected from laboratory for routine diagnostic purposes [19] sampled in an finger pricks using a lancet stick and applied to glass microscope EDTA tube before DNA extraction and negative for T. whipplei slides. Thick and thin blood smears were stained with Giemsa were tested by quantitative PCR b-actin as previously described and analyzed by microscopy for trophozoites and gametocytes [18]. Data from 18 patients with chronic asymptomatic T. whip- of malaria and Borrelia. If the malaria test result was negative, plei in the stool previously identified by our team [12] and 200 mL (3–4 drops) of whole blood was collected for DNA followed up for 4 years by one of us (D.R.) in consultation were also included for comparisons. Sixty blood specimens The first step of DNA extraction was done directly in the sampled in EDTA tubes for 4 years from these 18 patients were village dispensary using QIAamp kits (QIAGEN). The whole tested for T. whipplei using quantitative PCR as previously de- blood was digested with the supplied proteinase K. Then bind- scribed [19].
ing and washing of samples with Qiagen columns were per- Data were analyzed using EpiInfo soft- formed with an adapted 35-mL PVC manual pump (Fisher ware, version 3.4.1 (Centers for Disease Control and Preven- Scientific). Next, the columns with the bound and dried DNA tion). Nonparametric values were compared using the x2 test.
inside were stored at +4C before final elution performed later Statistical significance was defined as P values !.05. The cor- at the Unite des Rickettsies, Marseille, France. After repeated rected x2 test or the Fisher exact test was used when indicated.
516 • CID 2010:51 (1 September) • Fenollar et al
Figure 1.
Map illustrating the location of Senegal and the villages of Dielmo and Ndiop and the number of samples positive for Tropheryma whipplei in this study.
6.9 for 10 blood specimens received in our laboratory for rou- tine diagnostic purposes and sampled in an EDTA tube before From December 2008 to July 2009, 204 blood DNA extraction. The difference was not significant (P p .77) samples from 134 patients (62 male and 72 female patients) between these 2 ways of sampling.
were analyzed and included in this study; 103 (77%) were from The prevalence of T. whipplei–positive samples was 6.4% (13 Dielmo (391 inhabitants) and 31 (23%) were from Ndiop (313 of 204). The proportion of positive samples was 0 of 4 in inhabitants). Ninety (67.2%) were aged !10 years, including64 (47.8%) who were aged November, 3 (14.3%) of 21 in December; 7 (16.6%) of 42 in !5 years. No one died during the January, 0 of 7 in February, 0 of 25 in March, 3 (16.7%) of 18 Thirteen patients (Figures 1 and 2) tested positive for T. in April, 0 of 16 in May, 0 of 29 in June, and 0 of 42 in July.
whipplei according to our criteria. Furthermore, the b-actin The prevalence was significantly higher in children aged !10 gene amplification by quantitative PCR was highly positive for years (10 of 252) than in older people (3 of 452; P p .003).
all specimens, confirming the quality of the extracted DNA.
The prevalence was also significantly higher in Dielmo (12 [3%] Indeed, the mean number of cycles ( standard deviation of 391) than in Ndiop (1 [0.3%] of 313; P p .007) and in both [SD]) of the cycles was 25  2.85 for all of the blood specimens December and January (10 [15.9%] of 63) than during the directly sampled for DNA extraction in Senegal and 26.19  - other months (3 [2.1%] of 141; P ! .001).
Figure 2.
Summary of the 13 patients (coded as A–M) with Tropheryma whipplei–positive DNA on blood focusing on the seasonality of the bacteremia and the T. whipplei polymerase chain reaction data from saliva and stool performed at the time of systematic screening in April 2008 and 2009 butalso during an episode of diarrhea for 2 patients in January 2009.
T. whipplei Bacteremia • CID 2010:51 (1 September) • 517
Epidemiologic and Clinical Characteristics of the 13 Patients with Tropheryma whipplei Bacteremia and Results of the
Systematic Screening to Look for T. whipplei in Stool and Saliva Specimens, April 2009.
Clinical symptoms stool; for saliva Cough, rhinorrhea Negative; negative Headache, thirst, fatigue, sleep disorders Headache, fatigue, cough Headache, fatigue, cough, rhinorrhea Negative; negative Low appetite, diarrhea, cough Headache, fatigue, cough, rhinorrhea Headache, thirst, low appetite, fatigue, cough, Positive; negative rhinorrhea, sleep disorders Cough, rhinorrhea Positive; negative Sleep disorders, fatigue, headache, low Headache, sleep disorders, low appetite, Negative; negative thirst, fatigue, rhinorrhea Low appetite, cough, sleep disorders, vomiting NA, not available.
The main clinical characteristics of the patients who were 45 males and 41 females aged 7 months to 79 years (mean positive for T. whipplei are summarized in Table 1. Five patients SD, 913.3 years). The data are summarized in Table 3.
were male and 7 were female. Patients were aged 7 months to Among the 13 patients with T. whipplei bacteremia, stool sam- 49 years, with a mean age (SD) of 10.7  16.3 years, a median ples were analyzed for 10 (Figures 1 and 2). Among these 10 age of 3 years, and an interquartile range of 9 years. Their main patients, 4 also had a positive PCR result in stool samples, clinical manifestations were cough (reported by 10 patients), whereas 6 had negative results. Twenty-five patients with neg- thirst (8 patients), headache (7 patients), rhinorrhea (6 pa- ative T. whipplei PCR results in blood specimens had a positive tients), fatigue (5 patients), sleep disorders (5 patients), and stool T. whipplei PCR result at the time of the systematic screen- loss of appetite (3 patients). Only 1 patient presented withvomiting and 1 with diarrhea. All the patients were treated with Comparison of the Main Clinical Characteris-
tics of 13 Patients with Tropheryma whipplei Bacteremia
acetominophen; only 1 received antibacterial therapy with tri- and of 191 Patients without T. whipplei Bacteremia
methoprim-sulfamethoxazole. When compared with the 191episodes without T. whipplei bacteremia, cough and sleep dis- No. (%) of patients orders are significantly more frequent in patients with T. whip- plei DNA in their blood (P p .002 and P p .005, respectively) Six patients presented with new episodes of fever after T. whipplei bacteremia, nearly 2 months later for 5 patients and 7 months later for 1 patient. Other blood specimens were also available. All T. whipplei PCR assay results were negative for 5 patients. For 1 patient, T. whipplei DNA was only slightly de- tected in 1 of the 2 specific PCR assays 2 months later; however, 7 months later both PCR assay results were negative. None of the blood specimens from the 18 French patients who were chronic asymptomatic carriers of T. whipplei were positive for T. whipplei.
Analysis of stool and saliva specimens.
Of the 134 patients analyzed in this study, stool samples were also sampled for 86 at the time of systematic screening in Spring 2009, including Diarrhea was defined as 13 liquid stools per day.
518 • CID 2010:51 (1 September) • Fenollar et al
Comparison of Number of Stool Samples Positive for Tropheryma whipplei and Number Tested
for Each Age Group, April 2008 and 2009
No. of positive stool samples / no. tested (%), by date and region Data for April 2008 were extracted from our previous study [15].
ing. Seventeen were from Dielmo and 8 from Ndiop. Fifty-one identified cases of rickettsioses [20], Q fever [18], and recurrent patients with negative T. whipplei PCR results in blood speci- fever (unpublished data). To our knowledge, this study is the mens also had a negative stool T. whipplei PCR result at the first attempt to investigate the presence of T. whipplei in blood time of the systematic screening. Forty-two were from Dielmo specimens from patients with fever of uncertain origin. We and 9 from Ndiop. Overall, no significant differences were detected T. whipplei DNA in 6.4% of blood samples from pa- found between the presence of T. whipplei in stool samples at tients, primarily in children with fever in rural Senegal. We the time of the screening and T. whipplei bacteremia (P p .45).
believe that our findings are reliable. The validity of the data Furthermore, 33 patients tested for T. whipplei bacteremia had reported herein is based on strict experimental procedures and a negative stool PCR result in 2008. Among these 33 patients, controls, including rigorous positive and negative controls to 2 (1 from Ndiop and 1 from Dielmo) of 4 patients (1 from validate the test. In addition, each positive PCR result was Ndiop and 3 from Dielmo) who tested positive for T. whipplei confirmed by the successful amplification of an additional DNA in their blood also tested positive for T. whipplei in their stools in 2009, whereas 5 (4 from Ndiop and 1 from Dielmo) of 29 In a previous study, we found that 44% of asymptomatic patients (21 from Dielmo and 8 from Ndiop) who tested neg- children aged 2–10 years who lived in rural Senegal carried T. ative for T. whipplei in their blood had positive results for T. whipplei in their stool samples [15]. This screening was per- whipplei in their stool samples in 2009. This difference was not formed in April 2008. Herein, we confirm these results and significant (P p .18). Finally, 2 patients developed diarrhea in extent our data to the carriage of T. whipplei in saliva in older January 2009. T. whipplei DNA was detected in both stool sam- children and adults. The prevalence observed in individuals ples (Figure 1).
older than 10 years (18.2%) is lower than that in younger Among the 134 patients analyzed in this study, saliva samples individuals but higher than that observed in Europe [12, 14].
were collected for 51 at the time of the systematic screening in In addition, this is the first time, to our knowledge, that T. April 2009, including 33 females and 18 males aged 11 months whipplei in saliva has been detected in Senegal. The prevalence to 79 years (mean  SD, 15.4  18.5 years). The data are sum- of salivary carriage of T. whipplei (4%) is lower than that in marized in Table 4. Among the 13 patients with T. whipplei stool but is still higher than that previously observed in the bacteremia, saliva samples were analyzed for 7 patients from general population in France (0.2%) and even in sewer workers Dielmo. All test results were negative. Two patients from Ndiop (2.2%) [12, 19]. As in previous studies, the 2 individuals with with negative blood T. whipplei PCR results had a positive salivaT. whipplei PCR result at the time of the systematic screening.
Forty-two patients with negative blood T. whipplei PCR results Number of Saliva Samples Positive for Tropheryma
whipplei and Number Tested for Each Age Group, Spring 2009
also had a negative saliva T. whipplei PCR result at the time ofthe systematic screening. Thirty-one were from Dielmo and 11 No. of positive saliva samples / from Ndiop. No correlation was observed between T. whipplei no. tested, by region infection and positive saliva T. whipplei PCR results at the time of the screening (P p .74).
We are embarking on a large program of identification of the bacterial agents causing fever in these 2 villages. We have already T. whipplei Bacteremia • CID 2010:51 (1 September) • 519
Figure 3.
Proposal of a natural cycle for Tropheryma whipplei (Tw), including Tw bacteremia.
a PCR-positive saliva specimen also harbored T. whipplei in bacterium was detected in the bronchoalveolar lavage fluid of stool samples [12, 19].
a child with pneumonia in the United States and in 6 (2.85%) Although a seasonal variation for the detection of T. whipplei of 210 bronchoalveolar lavage fluid samples collected in patients DNA in blood specimens of patients with fever of uncertain with pneumonia in intensive care units in France, including 1 origin cannot be determined, significantly more cases were di- case in which T. whipplei was the only bacterium detected [7, agnosed in the beginning of the dry season in December and 8]. These overall data undoubtedly raise questions about the January (P ! .001). Villagers from Dielmo had a significantly role of T. whipplei in the genesis of respiratory diseases with- higher prevalence of T. whipplei bacteremia than those of out digestive involvement. It is currently difficult to locate Ndiop. Interestingly, malaria [16], recurrent fever [17], Q fever this bacteremia in the natural cycle of T. whipplei infection.
[18], and fleaborne spotted fever [20] also had higher preva- We propose several hypotheses (Figure 3).
lences in Dielmo than in Ndiop. The significant differences of We did not test afebrile control subjects from Senegal, but prevalence for these infectious diseases are commonly observed the presence of T. whipplei DNA in blood samples from ap- in neighboring places but currently remain unexplained. The parently healthy donors has been previously reported in a study main classic clinical manifestations linked to T. whipplei are with 1 positive sample of 174 (0.6%) [21]. In another study, arthralgia and diarrhea [1], but T. whipplei has been reported of 913 blood specimens tested by T. whipplei PCR for patients to involve nearly all organs [1]. Our hypothesis at the beginning with an excluded diagnosis of T. whipplei infection, none were of this work was that primo-infection resulting in gastroenteritis positive [19]. We may suspect that asymptomatic T. whipplei and fever [9] may be associated with bacteremia. However, we bacteremia is more common in rural Senegal, where the bac- were not able to link T. whipplei bacteremia to diarrhea or to terium is so common. This hypothesis should be addressed in the presence of T. whipplei in stool or saliva samples. However, the future because the lack of blood specimens from asymp- our 2 screenings were performed in the spring, whereas most tomatic individuals in Dielmo and Ndiop is the main limitation cases were observed in the winter.
of our study. However, the lack of bacteremia detected in French Our work raises the question of the existence of a link be- patients with chronic T. whipplei goes against this theory.
tween the presence of T. whipplei DNA in blood and the clinical We confirm that children in rural Senegal are infected with manifestations of infection. Herein, the major clinical sign of T. whipplei at an early age. Our preliminary results suggest that our patients was fever associated with cough. T. whipplei has T. whipplei may be associated with numerous undiagnosed in- been recently suggested as an agent of pneumonia [7, 8]. The fections with cough. We speculate that the distribution of T. 520 • CID 2010:51 (1 September) • Fenollar et al
whipplei is widespread and that the bacterium is an agent of 9. Raoult D, Fenollar F, Rolain JM, et al. Tropheryma whipplei in children with gastroenteritis. Emerg Infect Dis 2010; 16:776–782.
subtle, largely underestimated clinical manifestations, includ- 10. Dutly F, Altwegg M. Whipple's disease and Tropheryma whippelii. Clin ing respiratory infections.
Microbiol Rev 2001; 14:561–583.
11. Amsler L, Bauernfeind P, Nigg C, Maibach RC, Steffen R, Altwegg M.
Prevalence of Tropheryma whipplei DNA in patients with various gas-
trointestinal disease and in healthy controls. Infection 2003; 31:81–85.
We thank the Cre´dit Ministe´riel "Programme Hospitalier de Recherche 12. Fenollar F, Trani M, Davoust B, et al. Carriage of Tropheryma whipplei Clinique" 2009 (Recherche de Tropheryma whipplei comme agent de gastro- in stools of sewer workers and human controls, but not in monkeys ente´rite chez le jeune enfant).
and apes. J Infect Dis 2008; 197:880–887.
Cre´dit Ministe´riel "Programme Hospitalier de Re- 13. Maiwald M, Schuhmacher F, Ditton HJ, von Herbay A. Environmental cherche Clinique" 2009.
occurrence of the Whipple's disease bacterium (Tropheryma whippelii).
Potential conflicts of interest.
All authors: no conflicts.
Appl Environ Microbiol 1998; 64:760–762.
14. Schoniger-Hekele M, Petermann D, Weber B, Muller C. Tropheryma whipplei in the environment: survey of sewage plant influxes and sew-
age plant workers. Appl Environ Microbiol 2007; 73:2033–2035.
1. Fenollar F, Pue´chal X, Raoult D. Whipple's disease. N Engl J Med 15. Fenollar F, Trape JF, Bassene H, Sokhna C, Raoult D. Tropheryma whipplei in fecal samples from children, Senegal. Emerg Infect Dis 2009; 15:922–
2. Schneider T, Moos V, Loddenkemper C, Marth T, Fenollar F, Raoult D. Whipple's disease: new aspects of pathogenesis and treatment. Lan- 16. Trape JF, Rogier C, Konate L, et al. The Dielmo project: a longitudinal cet Infect Dis 2008; 8:179–190.
study of natural malaria infection and the mechanisms of protective 3. Altwegg M, Fleisch-Marx A, Goldenberger D, Hailemariam S, Schaffner immunity in a community living in a holoendemic area of Senegal.
A, Kissling R. Spondylodiscitis caused by Tropheryma whippelii. Schweiz Am J Trop Med Hyg 1994; 51:123–137.
Med Wochenschr 1996; 126:1495–1499.
17. Vial L, Diatta G, Tall A, et al. Incidence of tick-borne relapsing fever 4. Gubler JG, Kuster M, Dutly F, et al. Whipple endocarditis without overt gastrointestinal disease: report of four cases. Ann Intern Med 1999; 131:
in west Africa: longitudinal study. Lancet 2006; 368:37–43.
18. Mediannikov O, Fenollar F, Socolovschi C, et al. Humans and ticks as 5. Drancourt M, Fenollar F, Denis D, Raoult D. Postoperative panophthal- reservoirs of Q fever in rural Senegal. PLoS Negl Trop Dis 2010; 4:e654.
mitis caused by Whipple disease. Emerg Infect Dis 2009; 15:825–827.
19. Fenollar F, Laouira S, Lepidi H, Rolain J, Raoult D. Value of Tropheryma 6. Rickman LS, Freeman WR, Green WR, et al. Uveitis caused by Trophery- whipplei quantitative PCR assay for the diagnosis of Whipple's disease: ma whippelii (Whipple's bacillus). N Engl J Med 1995; 332:363–366.
usefulness of saliva and stool specimens for first line screening. Clin 7. Harris JK, De Groote MA, Sagel SD, et al. Molecular identification of Infect Dis 2008; 47:659–667.
bacteria in bronchoalveolar lavage fluid from children with cystic fi- 20. Socolovschi C, Mediannikov O, Sokhna C, et al. Rickettsia felis–associated brosis. Proc Natl Acad Sci U S A 2007; 104:20529–20533.
uneruptive fever, Senegal. Emerg Infect Dis 2010; 16:1140–1142.
8. Bousbia S, Papazian L, Auffray JP, et al. Molecular analysis of bron- 21. Misbah SA, Stirzaker D, Ozols B, Franks A, Mapstone NP. Anonymous choalveolar lavage in patients with pneumonia reveals the presence of survey of blood donors by polymerase chain reaction for Tropheryma Tropheryma whipplei. Emerg Infect Dis 2010; 16:258–263.
whippelii. QJM 1999; 92:61.
T. whipplei Bacteremia • CID 2010:51 (1 September) • 521

Source: https://www.cuso.ch/fileadmin/biologie_microbiologie/document/Fenollar_2010.pdf