Krisada Jongsakul

Scrub typhus infections poorly responsive to antibiotics in northern Thailand

BACKGROUND Rickettsia tsutsugamushi, the aetiological agent of scrub typhus, is common in Asia and readily infects visitors to areas where disease transmission occurs. Rapid defervescence after antibiotic treatment is so characteristic that it is used as a diagnostic test for R tsutsugamushi infection. Reports from local physicians that patients with scrub typhus in Chiangrai, northern Thailand responded badly to appropriate antibiotic therapy prompted us to do a prospective clinical evaluation and antibiotic susceptibility testing of human rickettsial isolates. METHODS The clinical response to doxycycline treatment in patients with early, mild scrub typhus in northern Thailand was compared with the results of treatment in Mae Sod, western Thailand. Prototype and naturally occurring strains of R tsutsugamushi were tested for susceptibility to chloramphenicol and doxycycline in mice and in cell culture. FINDINGS By the third day of treatment, fever had cleared in all seven patients from Mae Sod, but in only five of the 12 (40%) from Chiangrai (p < 0.01). Median fever clearance time in Chiangrai (80 h; range 15-190) was significantly longer than in Mae Sod (30 h; range 4-58; p < 0.005). Conjunctival suffusion resolved significantly more slowly in Chiangrai (p < 0.05). Antibiotics prevented death in mice infected by Chiangrai strains of R tsutsugamushi less often than after infection by the prototype strain (p < 0.05). Only one of three Chiangrai strains tested in cell culture was fully susceptible to doxycycline. INTERPRETATION Chloramphenicol-resistant and doxycycline-resistant strains of R tsutsugamushi occur in Chiangrai, Thailand. This is the first evidence of naturally occurring antimicrobial resistance in the genus Rickettsia.

Doxycycline and rifampicin for mild scrub-typhus infections in northern Thailand: a randomised trial.

BACKGROUND Some strains of scrub typhus in northern Thailand are poorly responsive to standard antirickettsial drugs. We therefore did a masked, randomised trial to compare rifampicin with standard doxycycline therapy for patients with scrub typhus. METHODS Adult patients with strictly defined, mild scrub typhus were initially randomly assigned 1 week of daily oral treatment with 200 mg doxycycline (n=40), 600 mg rifampicin (n=38), or doxycycline with rifampicin (n=11). During the first year of treatment, the combined regimen was withdrawn because of lack of efficacy and the regimen was replaced with 900 mg rifampicin (n=37). Treatment outcome was assessed by fever clearance time (the time for oral temperature to fall below 37.3 degrees C). FINDINGS About 12,800 fever patients were screened during the 3-year study to recruit 126 patients with confirmed scrub typhus and no other infection, of whom 86 completed therapy. Eight individuals received the combined regimen that was discontinued after 1 year. The median duration of pyrexia was significantly shorter (p=0.01) in the 24 patients treated with 900 mg daily rifampicin (fever clearance time 22.5 h) and in the 26 patients who received 600 mg rifampicin (fever clearance time 27.5 h) than in the 28 patients given doxycycline monotherapy (fever clearance time 52 h). Fever resolved in a significantly higher proportion of patients within 48 h of starting rifampicin (900 mg=79% [19 of 24], 600 mg=77% [20 of 26]) than in patients treated with doxycycline (46% [13 of 28]; p=0.02). Severe gastrointestinal events warranted exclusion of two patients on doxycyline. There were two relapses after doxycycline therapy, but none after rifampicin therapy. INTERPRETATION Rifampicin is more effective than doxycycline against scrub-typhus infections acquired in northern Thailand, where strains with reduced susceptibility to antibiotics can occur.

HIV-1 suppression during acute scrub-typhus infection.

BACKGROUND In HIV-1-infected individuals, viral load has been reported to rise transiently if an acute infection with another organism occurs. Our study was prompted by the unexpected finding that HIV-1 copy number fell during an acute infection with Orientia tsutsugamushi, the causative agent of scrub typhus. METHODS Serial HIV-1 viral load determinations were made in ten Thai adults with scrub typhus, who were not receiving antiretroviral therapy, and in five HIV-1-infected patients who had other infections (four malaria, one leptospirosis), during and after acute infections. Sera from HIV-1-infected patients with scrub typhus and from mice immunised with O. tsutsugamushi were examined for HIV-1-suppressive activity. FINDINGS Median viral load 3 days after admission was significantly lower in the scrub-typhus group than in patients with other infections (193% vs 376% of day 28 values, p=0.03). In four O. tsutsugamushi-infected patients HIV-1 RNA copy number fell by three-fold or more compared with day 28 values, and HIV-1 copy numbers were below the assay threshold in two patients with scrub typhus. Five of seven HIV-1 isolates from non-typhus patients with CD4 lymphocytes less than 200 cells/microL were syncytia-inducing variants, whereas all ten isolates from O. tsutsugamushi-infected individuals matched by CD4-cell count were non-syncytia inducing (p=0.03). Sera from an HIV-1-negative patient with scrub typhus had potent HIV-1-suppressive activity in vitro. Sera from typhus-infected mice inhibited HIV-1 syncytia formation and bound by immunofluorescence to HIV-1-infected lymphocytes. INTERPRETATION HIV-1-suppressive factors are produced during some scrub-typhus infection and should be investigated further in the search for novel strategies for the treatment and prevention of AIDS.

Blinded, Placebo-Controlled Trial of Antiparasitic Drugs for Trichinosis Myositis

There is no consensus on the benefits of treatment with any specific anthelminthic compound on muscle-stage trichinosis. A double-blind, placebo-controlled comparison was done of 3 antiparasitic drugs during an outbreak of trichinosis in Chiangrai Province, northern Thailand. Forty-six adults were randomized to receive 10 days of oral treatment with mebendazole (200 mg twice a day), thiabendazole (25 mg/kg twice a day), fluconazole (400 mg initially, then 200 mg daily), or placebo. All patients received treatment to eradicate adult intestinal worms. Trichinella spiralis infection was proved parasitologically in 19 (41%) of 46 patient and by serodiagnosis in all cases. Significantly more patients improved after treatment with mebendazole (12/12) and thiabendazole (7/7) than after treatment with placebo (6/12; P<.05) or fluconazole (6/12). Muscle tenderness resolved in more patients treated with thiabendazole and mebendazole than in those treated with placebo (P<.05). However, 30% of volunteers could not tolerate the side effects of thiabendazole. In summary, Trichinella myositis responds to thiabendazole and to mebendazole.

Decrease in human immunodeficiency virus type 1 load during acute dengue fever.

Rather than the expected increase in human immunodeficiency virus type 1 (HIV-1) load, there was transient suppression of HIV-1 replication during acute dengue infection in a 29-year-old Thai woman. Acute-phase (but not convalescent-phase) serum samples obtained from an HIV-1-uninfected patient with dengue fever reduced HIV-1 infectivity, as determined by a peripheral blood mononuclear cell assay, suggesting the possibility that HIV-1 replication is suppressed during acute dengue fever, as occurs during some cases of scrub typhus infection and measles.

Performance of a Dot Blot Immunoassay for the Rapid Diagnosis of Scrub Typhus in a Longitudinal Case Series

A rapid dipstick test for scrub typhus was prospectively evaluated in Chiangrai, northern Thailand. Sera from 162 patients with fever of unclear etiology were tested by a dot blot immunoassay using two different antigen concentrations. Dipsticks coated with lower concentration of antigen lacked sensitivity compared with the indirect immunoperoxidase test. Dipsticks with higher antigen concentration had increased sensitivity that was equivalent to that of the immunoperoxidase test. By increasing the antigen concentration on the dipstick, sensitivity increased from 67% to 100%, positive predictive value increased from 90% to 93%, and negative predictive value rose from 92% to 100%. The specificity of both antigen concentrations was 98%. This study establishes that scrub typhus can be confirmed serologically by use of a dipstick assay and that serodiagnosis can be effectively tailored to a target population.

Association of a Novel Mutation in the Plasmodium falciparum Chloroquine Resistance Transporter With Decreased Piperaquine Sensitivity

Background Amplified copy number in the plasmepsin II/III genes within Plasmodium falciparum has been associated with decreased sensitivity to piperaquine. To examine this association and test whether additional loci might also contribute, we performed a genome-wide association study of ex vivo P. falciparum susceptibility to piperaquine. Methods Plasmodium falciparum DNA from 183 samples collected primarily from Cambodia was genotyped at 33716 genome-wide single nucleotide polymorphisms (SNPs). Linear mixed models and random forests were used to estimate associations between parasite genotypes and piperaquine susceptibility. Candidate polymorphisms were evaluated for their association with dihydroartemisinin-piperaquine treatment outcomes in an independent dataset. Results Single nucleotide polymorphisms on multiple chromosomes were associated with piperaquine 90% inhibitory concentrations (IC90) in a genome-wide analysis. Fine-mapping of genomic regions implicated in genome-wide analyses identified multiple SNPs in linkage disequilibrium with each other that were significantly associated with piperaquine IC90, including a novel mutation within the gene encoding the P. falciparum chloroquine resistance transporter, PfCRT. This mutation (F145I) was associated with dihydroartemisinin-piperaquine treatment failure after adjusting for the presence of amplified plasmepsin II/III, which was also associated with decreased piperaquine sensitivity. Conclusions Our data suggest that, in addition to plasmepsin II/III copy number, other loci, including pfcrt, may also be involved in piperaquine resistance.