From Whipple Disease to Tropheryma whipplei Infection

(See the Major Article by Moter et al on pages 1089–97.) Whipple disease was first considered a metabolic disease, then an infectious disease, and finally a specific immunodeficiency to a common bacterium, Tropheryma whipplei, which also causes endocarditis, encephalitis, and acute infections. The existence of a new diagnostic test, by detection in urine, may help to differentiate diseases from carriages. Whipple disease was among the discoveries of the 20th century, and changed completely in the course of the 21st century. During the 20th century, the disease was first described by Georges Whipple, then a young doctor at John Hopkins Hospital, in a young doctor who died in a state of cachexia. Whipple found significant anomalies in the patient’s digestive tract, which were interpreted as a metabolic disorder. Later on, the efficacy of antibiotics and the analysis of samples by electron microscopy confirmed that it was a bacterial disease [1]. Subsequently, universal molecular amplification techniques revealed a bacterium, then unknown, in the digestive tract of patients with Whipple disease, whose established name was Tropheryma whipplei [2]. At the beginning of the 21st century, the culture of this bacterium was performed [3], which allowed the sequencing of its genome, thus establishing molecular tools much more specific than those based on the 16S gene, which was plagued by false positives [4]. Since then, the diffusion of these tools has shown that the bacterium associated with Whipple disease (T. whipplei) is ubiquitous. Whipple disease appears to have become a specific host disease, some genetic elements of which have recently been discovered [5], making the patient unable to eliminate a common bacterium. This specific immunodeficiency explains relapses and reinfections. This was the rationale for lifelong treatment. As a matter of fact, the occurrence of this bacterium in saliva and human feces is important, and accounts for 3% to >10% of samples, depending on the country, while the incidence of Whipple disease is 1 per million inhabitants, and mainly described in middle-aged white men [4]. Moreover, up to 50% of people in France have antibodies against T. whipplei. It appears that the prevalence of T. whipplei carriage depends on the circumstances and the country studied. Thus, in Europe, the sewage workers have much higher carriage rates than the general population, and carriage in rural Africa (such in Senegal), is much more important than in developed countries (eg, France, Germany, and Austria). It is likely that the disease has 2 major forms of transmission that are purely interhuman diseases related to fecal transmission in the poorest countries and saliva transmission in the wealthy countries [4]. Therefore, the use of polymerase chain reaction (PCR) on mucosal samples (saliva or stools) has a low predictive value, which may be increased when the estimated bacterial load is high (Table 1). Because of this, the work by Moter et al showing the detection of T. whipplei by PCR in urine, presented in this issue of Clinical Infectious Diseases [6], has the advantage of adding a new noninvasive sample to the existing repertoire of tests for the diagnosis of Whipple disease. The urine samples in this work are positive in 9 of 12 cases of Whipple disease, whereas they are negative in the controls, including people carrying T. whipplei in the feces without being sick. The authors conclude that this test can be useful in helping to diagnose Whipple disease. Polymerase Chain Reaction Test to Perform for the Various Clinical Manifestations of Tropheryma whipplei Infection—Proposed Place of Urinalysis Abbreviations: BALF, bronchoalveolar lavage fluid; CSF, cerebrospinal fluid; PCR, polymerase chain reaction. Polymerase Chain Reaction Test to Perform for the Various Clinical Manifestations of Tropheryma whipplei Infection—Proposed Place of Urinalysis Abbreviations: BALF, bronchoalveolar lavage fluid; CSF, cerebrospinal fluid; PCR, polymerase chain reaction. This new approach is indeed very important as it should also allow the diagnosis of other invasive diseases due to T. whipplei such as endocarditis, encephalitis, and arthritis. Really invasive Whipple disease can be localized at the beginning to the heart valves only without intestinal location. Tropheryma whipplei endocarditis is a cause of blood culture–negative endocarditis, relatively common in white men older than 50 years, with unexplained arthralgias [4]. Whipple disease may also present as a purely osteoarticular infection, at least initially, often aggravated by the immunosuppressive therapy prescribed for rheumatologic diseases, which may reveal a classical Whipple disease. In all of these cases of Whipple disease, a disease with tissue involvement, the detection of T. whipplei DNA in urine may be useful (eg, before starting an immunosuppressive treatment for joint inflammation in patients, without clear diagnosis). Moreover, the diagnosis of exclusively brain-localized Whipple disease (including relapses) is extremely difficult, especially since cerebrospinal fluid examinations are often negative, and often only cerebral biopsy (which is an extremely aggressive diagnostic procedure) can provide a diagnosis. As part of a diagnostic score, it is possible that the PCR performed on urine samples can increase the spectrum of diagnosed brain infections. It should be noted that the combination of obesity, cerebellar syndrome, and ocular involvement associated with cognitive impairment may be completely isolated with no other tissue involvement. Any additional noninvasive test is welcome in this context. Moreover, it seems that Whipple disease can present an acute infection in 3 types of circumstance: bronchopulmonary infection, bacteremia, and acute diarrheas. Indeed, the role of T. whipplei in bronchorespiratory infections is still poorly understood. Alongside pulmonary involvement during Whipple disease, T. whipplei was found in several bronchoalveolar lavage samples from 2 different teams, where it is cultivable. However, we were not able to confirm the lung involvement, when DNA was found in these conditions, despite the fact that the living bacterium could be isolated from a bronchoalveolar lavage sample. Under these circumstances, the presence of T. whipplei DNA in the urine may help confirm the invasiveness of T. whipplei in suspected pulmonary infections. Moreover, in a systematic study performed in Senegal, it was found that T. whipplei DNA could be retrieved from the blood of febrile children. The only symptom described in these children was the association of fever and cough. This study was performed in a small rural village, and we suspect that a small-cause outbreak was associated with the local baker; after his departure, the outbreak vanished. Finally, a systematic work performed in children with gastroenteritis, in Marseilles, showed that T. whipplei could be found, with rather high bacterial load, in children <2 years of age. Depending on the years, the frequency was higher, and could reach up to 15% of children presenting to the emergency department with relatively severe gastroenteritis. For the understanding and interpretation of these data, the presence of DNA in urine, as a noninvasive sample, would probably help to better understand and define whether it is necessary to prescribe specific treatment for diarrhea caused by T. whipplei. In practice, the new test presented by this team [6] can have major consequences in the understanding of T. whipplei infections, in the detection of relapses, in the primary diagnosis of classical Whipple disease (of course, it will be necessary to confirm this, at this stage, by a digestive biopsy), but also for T. whipplei infections without true, but localized Whipple disease (endocarditis or isolated encephalitis). This test can also help us to better understand the role of T. whipplei in isolated acute infections, such as pneumonitis or gastroenteritis. The T. whipplei infection spectrum is incompletely known, as the bacterium as been described only in the 21st century, and any new test is welcome to increase our capabilities to expand our knowledge. Potential conflicts of interest. Author certifies no potential conflicts of interest. The author has submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.