Sir, Whipple’s disease is a chronic infection caused by Tropheryma whipplei. In a prospective study, the CNS was found to be involved in 38.5% of the cases. Treatment is not always successful. – 6 Co-trimoxazole (trimethoprim/sulfamethoxazole) has been reported to be significantly more effective than tetracycline; however, resistance to co-trimoxazole has also been observed. – 6 More recently, there have even been reports describing resistance of CNS infections to treatment with ceftriaxone, a bactericidal antibiotic penetrating the blood–brain barrier. In 2002, a patient presented with diarrhoea, weight loss to 64 kg, anaemia (8.4 g/dL haemoglobin) and erythrocyte sedimentation rate of 33 mm/h. The medical history revealed relapsing arthritis since 1991 and pericarditis necessitating pericardial resection in 1993. Gastrointestinal biopsies disclosed periodic acid-Schiff (PAS)-positive macrophages typical of untreated Whipple’s disease in the mucosa of the duodenum and the ileum and in the submucosa of the colon. The patient had no cerebral symptoms. However, microscopic examination of centrifuged CSF obtained by spinal puncture showed a PAS-positive macrophage typical of Whipple’s disease (histopathology carried out by Dr Reinhard Golz, Wuppertal), and the PCR to T. whipplei in the CSF was positive. The patient was admitted to a prospective treatment trial, as reported in Feurle et al. He was randomized to 2 g of ceftriaxone infused intravenously once daily for 2 weeks, followed by oral co-trimoxazole at a dosage of 160/300 mg twice daily for 12 months. While the patient recovered from all signs and symptoms of Whipple’s disease, the PCR for T. whipplei remained positive in the CSF for 5.5 years despite additional treatment with 1 g of meropenem infused intravenously thrice daily for 2 weeks followed by cotrimoxazole for another year. After a further year of cotrimoxazole, the CSF was still positive in the T. whipplei PCR, while the patient was receiving this treatment. PAS-positive macrophages in duodenal mucosal biopsies and in the CSF had disappeared. CT and magnetic resonance imaging of the brain did not reveal any structural abnormality. At this time, the patient had no symptoms of systemic or cerebral Whipple’s disease. The presence of T. whipplei in the CSF was confirmed by sequence analysis of the amplification product, and the viability was established by culture in MRC5 fibroblasts and in axenic medium. In vitro susceptibility of the cultured strain to ceftriaxone, meropenem, tetracyclines and co-trimoxazole is shown in Table 1. The risk of this patient developing symptomatic cerebral disease seemed unpredictable. After obtaining written informed consent for an individual treatment attempt, this patient was treated with chloroquine and tetracycline, a combination suggested previously. Chloroquine (or hydroxychloroquine) enhances antimicrobial activity of tetracyclines by raising the pH of the phagolysosomes within macrophages. Minocycline was selected as the tetracycline in this case as this compound has been reported to cross the blood–brain barrier well. Pollock et al., taking advantage of this pharmacokinetic property, were probably the first to treat a patient with cerebral Whipple’s disease with minocycline. They did not combine it with chloroquine. Others prefer the combination of doxycycline and chloroquine. In the present study, the dosage of chloroquine, based on a body weight of 83 kg, was 1725 mg of chloroquine phosphate in divided doses orally on day 1, and 575 mg on the second to the fourth day, followed by one 250 mg chloroquine phosphate tablet once daily for 45 days. A serum concentration of 165 mg/ L chloroquine (therapeutic range 20 to 200 mg/L) was obtained. Minocycline was given at a dosage of one 100 mg minocycline tablet twice daily for 45 days. This dosage had to be reduced to 100 mg daily from day 5 to day 8 because of vertigo. Vestibular toxicity manifesting as dizziness, ataxia and nausea is a common side effect of this tetracycline. The plasma concentration of minocycline, determined by HPLC, was 2.1 mg/L. The serum and respective plasma concentrations were determined 5 weeks after onset of treatment and 14 h after the last medication. The PCR for T. whipplei in the CSF was negative, both 12 and 24 months later. The patient has been asymptomatic now for more than 9 years after the initial examination and treatment. Obviously, this new treatment option, found to be efficacious in a single patient, should be evaluated in a prospective trial. The regimen is attractive as the combination of minocycline and chloroquine can be given orally. However, vestibular side effects (occurring in .70% of women taking 100 mg of minocycline every 12 h) will render any study difficult. Our male patient had to reduce the dose of minocycline for several days.