Abstract
Fluoroquinolones are the drugs of choice for the treatment of typhoid fever caused by multidrug-resistant Salmonella enterica serovar Typhi (MDR-ST) infections. Decreased susceptibility of S. Typhi isolates to fluoroquinolones is emerging as an important problem in South Asian countries such as India, Vietnam, and elsewhere (Parry, 2004; Rahman et al., 2006; Renuka et al., 2005). Nalidixic acid resistance is useful as a surrogate marker of decreased susceptibility to fluoroquinolones (Hakanen et al., 1999). We have previously reported the association of nalidixic acid-resistant S. Typhi (NARST) infection with prolonged duration of illness at presentation, increased clinical severity of illness and complications (Kadhiravan et al., 2005). Here, we present the results of the exploratory analysis of the data regarding the association of drug resistance with fever clearance time.
The study group comprised 60 consecutive patients, including children, with blood culture-confirmed typhoid fever, treated at a tertiary care hospital in northern India, followed up prospectively. The patient characteristics and study methods have been described previously (Kadhiravan et al., 2005). Fever clearance time was defined as the time from the initiation of antibiotic therapy to the first instance when body temperature fell below 37.5 °C and remained so continuously for at least 48 h. In the present analysis, 51 patients in whom the fever clearance time could be assessed are considered. The effect of infection with NARST and MDR-ST on fever clearance time was assessed by Kaplan–Meier survival analysis, to display the fever clearance distributions of the groups (Bland & Altman, 1998). The log-rank test was applied to evaluate the equality of the fever clearance distributions (Bland & Altman, 2004). Further, a multivariable Cox proportional-hazards regression model was used to evaluate the relationship between NARST infection and the fever clearance time, adjusting for other potential confounding variables. All tests were two-sided, and a P <0.05 was considered statistically significant. All analyses were performed using a statistical software package (SPSS for Windows, version 10.0.1).
The mean age of the study group was 16±9 years, and one-third of the patients were female [17 (33 %) patients]. The isolate was resistant to nalidixic acid in 41 (80 %) patients, and 18 (35 %) patients had MDR-ST infection. The median time to fever clearance was 3 days [interquartile range (IQR) 2.75–5 days] in patients with nalidixic acid-susceptible S. Typhi (NASST) isolates and was 5 days (IQR 3.9–6.1 days) in patients infected with NARST. The difference between the NARST and NASST groups in the distribution of fever clearance time was statistically significant (Fig. 1a; P=0.028). At any given point of time, fever clearance was about twice more likely to occur in patients with NASST as compared to those with NARST infection (unadjusted hazard ratio for NASST vs NARST infection, 2.19; 95 % confidence interval 1.05–4.61). However, there was no significant difference in the distribution of the fever clearance time between patients with MDR-ST infection and the rest of the study group (Fig. 1b; P=0.948).