Skip to main content
Download PDF

Doxycycline vs azithromycin in patients with scrub typhus: a systematic review of literature and meta-analysis

BMC Infectious Diseases volume 23, Article number: 884 (2023) Cite this article

Abstract

Introduction

Scrub typhus is a bacterial mite-borne disease associated with poor clinical outcomes if not treated adequately. The study aimed to compare the time to defervescence, clinical failure, mortality and treatment-related adverse effects of two common drugs (doxycycline and azithromycin) used for its treatment.

Methodology

This was a systematic review and meta-analysis. All studies up to 20.03.2023 were screened for eligibility in Pubmed and Embase using a search string containing terms related to scrub typhus, doxycycline and azithromycin. After two phases of screening, all comparative studies where doxycycline and azithromycin were used to treat scrub typhus were included. The studies were critically appraised using standardised tools, and a meta-analysis was performed for time to defervescence (primary outcome), clinical failure, mortality and treatment-related adverse effects.

Results

Of 744 articles from two databases, ten were included in the meta-analysis. All but two studies had a high risk of bias. The meta-analysis for time to defervescence had a high heterogeneity and did not show any significant difference between doxycycline and azithromycin arms [Mean difference of -3.37 hours (95%CI: -10.31 to 3.57), p=0.34]. When the analysis was restricted to studies that included only severe scrub typhus, doxycycline was found to have a shorter time to defervescence [mean difference of -10.15 (95%CI: -19.83 to -0.46) hours, p=0.04]. Additionally, there was no difference between the two arms concerning clinical failure, mortality and treatment-related adverse effects.

Conclusion

The current data from studies with a high risk of bias did not find statistically significant differences in clinical outcomes between doxycycline and azithromycin for scrub typhus

Peer Review reports

Introduction

Acute undifferentiated febrile illness (AUFI) is a broad syndromic term that includes infections with a fever of short duration without specific localisation [1]. It is a common cause of morbidity and mortality in tropical countries [1]. Scrub typhus, a mite-borne infection caused by Orientia tsutsugamushi, is a leading cause of AUFI in Asia [2]. The disease was initially thought to be restricted to an imaginary triangle called the ‘tsutsugamushi triangle’ (Indian subcontinent, South-East Asian countries, China, Korea, Japan, and Northern Australia) [3]. However, recent cases have also been reported in Peru and Kenya [3]. Despite the high burden, scrub typhus is neglected, with limited high-quality studies available to inform treatment decisions [4]. In a systematic review from India, scrub typhus was the aetiology in a quarter of hospitalised patients with AUFI [5]. Around 20% of scrub typhus patients had severe manifestations and required intensive care [5]. The median mortality of scrub typhus is 6% in untreated patients, but it falls to 1.4% in adequately treated patients [2]. In a systematic review of the treatment landscape, doxycycline and azithromycin were the most common antibiotics used for management [4]. Indian Guidelines recommend either drug to treat scrub typhus without any clear preference [6]. With increasing awareness and improved diagnostics, many comparative studies on the management of scrub typhus have recently been published. There is a need for a meta-analysis combining the results of all published studies, including the recent ones, to identify the drug with better clinical outcomes. This study aimed to compare time to defervescence, clinical failure, mortality and treatment-related adverse effects between doxycycline and azithromycin in patients of all ages and sexes diagnosed with scrub typhus.

Methodology

Search strategy

This systematic review and meta-analysis was conducted after Prospero registration (CRD42023409680). The search string was developed using relevant search terms for population (febrile patients diagnosed with scrub typhus) and intervention/comparator (doxycycline and azithromycin) (Supplementary Table 1).

Study selection

Articles in all languages from the two databases (Pubmed and Embase) up to 20/03/2023 were included. The detailed eligibility criteria have been tabulated (Supplementary Table 2). Briefly, randomised controlled trials (RCTs) and non-randomized clinical trials (nRCTs) on scrub typhus, where doxycycline and azithromycin were used for the treatment of individuals of any age and sex (including pregnant individuals) diagnosed with scrub typhus were included. Studies on non-human subjects and those focusing exclusively on coinfections were excluded. Case reports, case series, reviews, systematic reviews, conference abstracts and letters to the editor were also excluded.

The articles were retrieved from the two databases after removing the duplicates. Two authors (NG and CB) independently screened the titles and abstracts for eligibility. After the initial screening, full-length articles were retrieved for screening by the same two authors and the articles that met the eligibility criteria were included in the final analysis. The third author (SVDB) was consulted for conflicts between the two reviewers.

Outcome measures

The primary outcome measure of the study was time to defervescence. The secondary outcome measures were clinical failure, mortality and treatment-related adverse effects. Time to defervescence was defined as the time taken for the temperature to go below a cut-off defervescent point. This was only considered if the fall in temperature was sustained for more than 24 hours. Clinical failure was defined as the proportion of patients not attaining defervescence in two or three days after initiation of treatment. The all-cause mortality during the follow-up period in each treatment arm was also noted. In those studies where all-cause mortality was unavailable, scrub-typhus-related mortality was recorded. Furthermore, adverse effects in each arm attributed to either doxycycline or azithromycin were recorded.

Data extraction

The following population-related details were extracted from studies: author details, type of study, diagnostic modality, age group of the included patients (adult or paediatric) and disease severity. Severe scrub typhus was defined as the presence of any organ dysfunction. The number of patients with severe scrub typhus was recorded in each study. The dose, duration, and route of doxycycline and azithromycin were recorded. The definition of fever and how the temperature was measured were recorded for each study. The mean or median time to defervescence of fever in hours, proportion of patients with clinical failure, proportion of patients who died and proportion of patients with treatment-related adverse effects in each treatment arm were noted.

Critical appraisal of literature

Each included article was appraised using the Cochrane Risk of Bias tool for RCTs and the ROBINS-1 tool for nRCTs (including analytical observational studies) [7, 8].

Data analysis

Results from all the studies were compared to determine differences in the outcome measures between doxycycline and azithromycin. The mean difference of time to defervescence was pooled using a random-effect model (inverse variance method), and the results were represented by a point estimate with a 95% confidence interval. In the studies where mean time to defervescence and standard deviation were not mentioned, they were indirectly calculated from the median and interquartile range (or range) using the method suggested by Wan et al. [9]. The meta-analysis results were stratified according to the study type, age groups and severity. The risk ratio for clinical failure, mortality and adverse events was pooled using a random-effect model (Mantel Haenszel method), and the results were represented by a point estimate with a 95% confidence interval. The heterogeneity across studies for all the outcomes was tested using Tau2, Chi2 and I2 tests. An I2 of more than 60% was taken as significant heterogeneity. Review Manager (version 5.3, Cochrane Nordic, Copenhagen, Denmark) was used for the meta-analysis.

Results

Study selection

A total of 219 articles were retrieved from Pubmed, and 727 were retrieved from Embase. After duplicate deletion, a total of 744 articles were included. After the title abstract screening, 29 articles were included for full-text screening. Nineteen papers were excluded during the full-text review. Ten articles were finally included for data extraction and analysis (Fig. 1).

Fig. 1
figure 1

PRISMA diagram showing the two screening phases and the number of included and excluded articles

Full size image

Single-arm studies (not included)

Eight studies had data on time to defervescence for either azithromycin or doxycycline but not both [10,11,12,13,14,15,16,17]. They were, therefore, not included in the final meta-analysis. The average time to defervescence in these studies ranged from 24 to 52 hours for doxycycline and 24 to 43 hours for azithromycin (Supplementary Table 3).

Baseline characteristics of included studies

Ten studies with 2856 patients were included in this analysis, with sample sizes ranging from 57 to 780 [18,19,20,21,22,23,24,25,26,27] (Table 1). Seven of the ten studies were published during or after 2021. Most studies were reported from India (n=5, 50%). There were five RCTs [18, 21, 25,26,27]. Four studies included only paediatric patients (less than 14-18 years) [21, 22, 24, 25]. The remaining six incorporated either adults or individuals of any age. Scrub typhus was defined primarily based on the presence of eschar and/or serological tests (Weil Felix test or, rapid diagnostic tests or Enzyme-linked immunosorbent assay or Immunofluorescence assay). Additional details on the diagnostic tests used for diagnosis have been summarised in Supplementary Table 4. Six studies excluded the use of antibiotics (active for scrub typhus) before presentation (Supplementary Table 4). Only one study explicitly mentioned that co-infections were ruled out [24]. Four studies exclusively focussed on patients with mild scrub typhus [21, 25,26,27]. Three studies focused entirely on severe scrub typhus patients [18, 19, 23]. The other three studies had severe scrub typhus ranging from 16% to 51% [20, 22, 24].

Table 1 Comparative studies for time to defervescence or mortality on doxycycline vs. azithromycin for scrub typhus
Full size table

Outcomes listed in the included studies

All the included studies had time to defervescence as either a primary (n=3) or a secondary outcome (n=7) (Supplementary Table 5). Clinical failure (as defined in the methodology) was reported in nine studies. It was reported as the primary outcome in five studies. Mortality was mentioned in six studies [18,19,20, 23, 26, 27]. Only one of the studies mentioned mortality as the primary outcome. Adverse drug reactions were reported in five studies.

Definitions of outcomes

The definition of time to defervescence differed across the studies. The period of sustained temperature fall below a pre-defined cut-off (measured by oral or axillary thermometer) required to define defervescence varied from 24 to 72 hours across studies (Table 2). Only five out of ten studies mentioned that time to defervescence was calculated after explicitly excluding the use of antipyretics. Only five studies mentioned the frequency of temperature measurement to see when the temperature fell below the cut-off point. Except for one study, which monitored temperature 24 hourly, the rest of the four studies measured temperature every two to four hours (Table 2). Of the nine studies that reported clinical failure, seven reported failure to defervesce at two days. The two remaining studies reported failure to defervesce at three days (Table 2). Four studies defined mortality in their methodology. Two studies reported scrub-typhus-related mortality, while the other two reported all-cause mortality [18,19,20, 23]. Two additional studies reported mortality but did not define them [26, 27].

Table 2 Definition of outcomes for the studies included in the meta-analysis
Full size table

Details of the interventions

Doxycycline and azithromycin were used at similar dosages. Doxycycline in adult studies was used at 100 mg twice daily, whereas in paediatric patients, it was given at 2.2 mg/kg twice daily. Similarly, azithromycin was dosed at 500 mg (in adults) or 10 mg/kg (in children) once daily. Loading dose was given in three studies (Table 3). Most studies used oral formulations of both drugs except for the two studies on severe scrub typhus, where intravenous formulations were used. Additionally, azithromycin was used in intravenous formulation in the study by Hwang et al. The duration of treatment for doxycycline ranged from 5 to 8 days, while the course of treatment for azithromycin ranged from 1 to 7 days (Table 3). The most common duration for both drugs was seven days.

Table 3 Outcomes according to intervention in studies included in the meta-analysis
Full size table

Effect estimates in the individual studies

The time to defervescence with doxycycline ranged from 12 to 96 hours, while it was 21 to 96 hours for azithromycin (Table 3). The clinical failure rates (Day 2-3) varied from 4-41% in doxycycline and 2-70% in azithromycin (Table 3). When the cut-off for attaining defervescence was taken as five days to define failure, the failure rates in both arms varied between 0 and 7%. Mortality with doxycycline and azithromycin ranged from 0 to 10.9% and 0 to 12.2%, respectively (Table 3).

Critical appraisal of the included studies

The critical appraisal of the five RCTs showed that only one had a low risk of bias overall. The rest of the four had a high risk of bias (Fig. 2). The critical appraisal of the n-RCTs showed that all of them had a moderate or high risk of bias (Fig. 2).

Fig. 2
figure 2

Risk of bias analysis using (A) Cochrane Risk of Bias tool for randomised controlled trials and (B) ROBINS-1 tool for non-randomised clinical trials (including observational studies). Abbreviation: RCT- Randomised controlled trials, nRCTs- non-Randomised controlled trials, D- Domains of bias. *Colour grading- Red- High risk of bias, Yellow- Moderate risk of bias and Green- Low risk of bias. D1 to D7 stands for the various domains of critical appraisal

Full size image

Time to defervescence meta-analysis

Nine studies were included in the meta-analysis for the time to defervescence. The study by Hwang et al. could not be included in the time to defervescence meta-analysis as it just reported the median without reporting the standard deviation, range or inter-quartile range [20]. The meta-analysis of mean time to defervescence between doxycycline and azithromycin showed very high heterogeneity and did not show any difference [Mean difference of -3.37 hours (95%CI: -10.31 to 3.57), p=0.34] (Fig. 3). There was no difference between the two arms when the analysis was restricted to randomised controlled trials (Fig. 3). No significant differences in the adult and paediatric subgroups were noticed concerning the mean time to defervescence (Supplementary Fig. 1). An additional sensitivity analysis was done where the studies that focussed exclusively on mild or severe scrub typhus were included. The meta-analysis of studies which included severe patients showed that the mean time to defervescence with doxycycline was significantly lower when compared to azithromycin [mean difference of -10.15 (95%CI: -19.83 to -0.46) hours, p=0.04] (Fig. 4). The I2 for heterogeneity was 47%. Similar results were obtained when the standardised mean difference was taken as the effect measure of choice (Supplementary Fig. 2).

Fig. 3
figure 3

Meta-analysis of doxycycline vs. azithromycin for mean time to defervescence classified according to study type

Full size image
Fig. 4
figure 4

Meta-analysis for time to defervescence with studies stratified according to severity

Full size image

A separate subgroup analysis was conducted between studies that used loading dose and studies that did not. Although heterogeneity was decreased in those who received loading doses, no significant difference between doxycycline and azithromycin was noted in the two subgroups (Supplementary Fig. 3).

Clinical failure meta-analysis

The meta-analysis comparing clinical failure rates (Days 2-3) in nine included studies did not show any difference between the two arms [RR-0.78 (95%CI: 0.58-1.05), p=0.1] (Fig. 5). On sub-group analysis, the doxycycline arm showed a lesser risk for clinical failure on day two, but the difference was not significant [0.76 (95%CI: 0.55-1.05), p=0.1]. When failure to defervesce was evaluated on Day 5, the difference in clinical failure rates was insignificant between both arms [RR- 0.78 (95%CI: 0.16-3.69), p=0.75] (Supplementary Fig. 4).

Fig. 5
figure 5

Meta-analysis of doxycycline vs. azithromycin showing clinical failure rates on Days 2 to 3

Full size image

Mortality meta-analysis

Two of the six studies reported no mortality in either arm. Since the meta-analysis of dichotomous variables does not consider studies with no outcome events, only four studies were used for mortality meta-analysis [26, 27]. Mortality in the two arms was not significantly different [RR: 0.82 (95%CI: 0.54-1.24), p=0.34] (Fig. 6). There was no significant heterogeneity when studies were evaluated to calculate the risk ratio for mortality.

Fig. 6
figure 6

Meta-analysis of doxycycline vs. azithromycin for mortality

Full size image

Treatment-related adverse effects meta-analysis

Of the five studies that reported treatment-related adverse effects, one reported only those adverse effects that were Grade 3 or higher [18]. In the study by Phimda et al., adverse effects of patients diagnosed with either scrub typhus or leptospirosis were reported without making any distinction between them [26]. Therefore, only three studies were included in the meta-analysis. Both arms had similar treatment-related adverse effects [RR-1.37 (95%CI: 0.64-2.94). p=0.33] (Supplementary Fig. 5).

Because fewer than ten studies were in the meta-analysis, funnel plots were not created for any outcomes.

Discussion

Orientia tsutsugamushi is a gram-negative organism belonging to the Rickettsiaceae family [28]. Owing to its intracellular nature, scrub typhus can only be treated by antibiotics that accumulate in large quantities within the target cells. The first antibiotic used to treat scrub typhus was chloramphenicol [28]. Due to its significant toxicity profile, it has been predominantly replaced by doxycycline and azithromycin [18]. In this systematic review and meta-analysis, after screening a total of 744 articles published before March 2023, 10 articles were included in the final meta-analysis. This study included seven additional comparative articles not included in the previous meta-analyses [28, 29]. There was no significant difference between doxycycline and azithromycin concerning the time to defervescence, clinical failure, mortality and treatment-related adverse effects.

While both drugs act by inhibiting protein synthesis, there are significant differences in their mechanism of action [29]. Doxycycline acts on the 30S ribosomal unit and inhibits protein synthesis, which might lead to a quicker reduction in the bacterial load. However, azithromycin, acting on the 50S ribosomal unit, may have a longer-lasting effect on bacterial replication, potentially explaining the comparable clinical outcomes in terms of time to defervescence [29]. The two antibiotics also differ according to the WHO Access Watch and Reserve (AWARE) antibiotic classification [30]. While doxycycline is classified as an ‘Access’ group antibiotic, azithromycin is classified into the ‘Watch’ group [31]. This also has a bearing on the cost and availability of these drugs.

Time to defervescence was selected as the primary outcome in this SR as it was amongst the commonest primary outcomes in the selected studies. A longer time to defervescence leads to unnecessary escalation of antibiotics. Since fever is the most important presenting feature of scrub typhus, clinicians keep adding additional antibiotics until defervescence in routine clinical practice. In a randomised controlled trial from India, the average number of antimicrobials prescribed to patients with AUFI was 2.5 [32]. Overuse of antimicrobials can fuel antimicrobial resistance, and since scrub typhus endemic areas are also faced with significant antimicrobial resistance, choosing an intervention that decreases antimicrobial overuse is pertinent. Also, increased time to defervescence is associated with more extended hospital stays, thereby increasing the hospital burden and risk of acquisition of nosocomial infections. We chose to analyse time-to-defervescence as a continuous outcome (mean difference) and not a time-to-event analysis, as almost all study participants who contributed some period of time ended in the event (defervescence). Besides, the log hazard ratio required for conducting a time-to-event meta-analysis was unavailable in all the studies.

Nine studies with moderate to high risk of bias used in meta-analysis of time to defervescence revealed significant heterogeneity and no significant difference between the two arms, even when the analysis was restricted to RCTs. Since observational studies tend to exaggerate the true effect estimates, we did not attribute much importance to the sub-group analysis of nRCTs showing shorter defervescence with doxycycline. To reduce the heterogeneity in the primary outcome measure, we conducted a sensitivity analysis of studies that included only severe scrub typhus. The heterogeneity decreased substantially in this sensitivity analysis; doxycycline performed better than azithromycin.

All except one RCT by Varghese et al. had a high risk of bias in the quality assessment [18]. Despite the low risk of bias in the trial, the temperature measurement strategy (24 hourly) was not ideal for detecting differences of less than 24 hours between the two arms. The study also reported a wide interquartile range in both arms (4 to 20 days in doxycycline and 4 to 28 days in the azithromycin arms). It is unusual to see fever persisting for up to 28 days in scrub typhus. It is possible that these critically ill patients might have had other causes of fever (thrombophlebitis, hospital-acquired infection, etc) after the actual defervescence was obtained. The indirectly calculated mean and standard deviation in Varghese et al. ranged from 224 to 296 hours and 286 to 447 hours, respectively. The mean in other studies ranged from 24 to 65 hours, while the standard deviation ranged from 10 to 39 hours. Despite a substantial sample size, the weight of the study in this meta-analysis by inverse variance approach was meagre due to unusually high variance. As an additional analysis, we calculated a pooled standardised mean difference to counter the significant variance seen in Varghese’s study (Supplementary Fig. 2). In this analysis, the weight of the Varghese study is significantly higher (16.3%) as compared to the meta-analysis calculating mean-difference (2.3%) (Supplementary Fig. 2). Despite the increased weight, the overall results remain unchanged. Doxycycline has a slightly faster time to defervescence in patients with severe disease but not in mild patients. A standardised mean difference of 0.41 in patients with severe disease would indicate a small to moderate effect size.

In the study by Varghese et al., the combination regimen of doxycycline and azithromycin combination regimen showed better composite outcomes (28-day mortality, persistent complications at Day 7 and persistent fever at Day 5) when compared to doxycycline or azithromycin alone [18]. It must be noted that the combination regimen did not show any significant difference when mortality or persistence of fever were evaluated separately. Lesser persistent complications possibly drove the favourable outcomes in the combination arm. These included persistently high bilirubin (≥2mg/dl) or creatinine (≥2mg/dl) at day 7. It can be argued that these are reversible biochemical changes with limited impact on clinically relevant outcomes. Combination therapy was not included as one of the interventions in this meta-analysis due to a lack of similar RCTs focussing on its use.

Traditionally, azithromycin is considered unsafe in children (below eight years of age), as doxycycline use was deemed unsafe because of the problems with dental staining [29, 30]. Recent evidence, however, suggests that doxycycline can be used for short courses in these populations [29]. Four comparative studies in the paediatric population were reported in our meta-analysis. No difference between the two drugs was noted in the paediatric subpopulation. It is also essential to discuss scrub typhus in pregnancy. Of the 33 pregnant patients with scrub typhus in a study from India, poor foetal outcomes (intra-uterine deaths, spontaneous abortion) were observed in more than half of the patients [33]. All patients in this study were treated with azithromycin. Traditionally, azithromycin is considered the safer choice, and as a result, comparative studies between doxycycline and azithromycin in this subgroup have not been reported. With increasing evidence on the safety of doxycycline in pregnancy, it is anticipated that comparative studies between doxycycline and azithromycin will be available.

Clinical failure in most studies was defined as the failure to defervesce in a pre-specified amount of time. Most studies used two or three days as the cut-off. Three studies reported a cut-off at five days and the frequency of patients failing to defervesce at two days. For homogeneity, the cut-off for clinical failure was taken as two or three days. Some studies, such as the one by Varghese et al., used a composite outcome (all-cause mortality at day 28, persistent complications at day 7 and persistent fever at day 5), with one of the outcomes being failure to defervesce [18]. Since the definition of composite outcomes varied across studies, we focussed on these individual outcomes. There was no difference in clinical failure rates between the two drugs. These results of doxycycline achieving somewhat faster defervescence on Day 2 but not Day 3 or 5 can be partially explained by the longer time azithromycin takes to reach the steady state [34].

The meta-analysis for mortality showed no significant difference between doxycycline and azithromycin. This finding is consistent with previous studies and suggests that both antibiotics are equally effective in preventing fatal outcomes in scrub typhus patients. It must be noted that this analysis had minimal heterogeneity and a non-significant trend towards possible better mortality outcomes with doxycycline. Since the highest number of events were recorded in the study by Varghese et al., this study drove the results of the meta-analysis. Mortality was taken as the secondary and not as the primary outcome as the number of outcome events in all treated patients (doxycycline or azithromycin) was anticipated to be low [2]. Doxycycline and azithromycin differ in their side-effect profile, with the former associated with more gastrointestinal side effects and the latter associated with cardiac side effects, such as QT interval prolongation [29]. The meta-analysis indicated that treatment-related adverse effects were similar between the two drugs. This is an important consideration when choosing an antibiotic treatment, as safety and tolerability play a crucial role in patient management. In the study by Varghese et al., neither the doxycycline nor azithromycin arm had any Grade 3 or higher treatment-related adverse event.

It is essential to understand whether the efficacy of the two drugs is comparable or whether one is better than the other. The study results show that the efficacy of both drugs is comparable for clinically relevant outcomes. Historically, doxycycline was considered the drug of choice. Given the current study's findings, azithromycin can be considered a worthy alternative. In clinical settings, individual case-based decisions can be taken based on the dose frequency, availability, cost and patient profile. Although recent studies have shown minimal risk of doxycycline use in pregnancy and children, some clinicians prefer using azithromycin. Doxycycline can be used in patients with old age and myocarditis, where clinicians want to avoid azithromycin due to its effect on QT prolongation [35]. In a retrospective cohort study of older adults with pneumonia, myocardial infarction was higher in azithromycin when compared to other antibiotics [35]. In a study that compared Clostridium difficile infection (CDI) between doxycycline and azithromycin, CDI rates were lower with doxycycline [36]. Therefore, doxycycline may be a better choice for individuals with a high risk of CDI. Although doxycycline is the preferred choice for numerous infections involving the central nervous system, some pharmacological studies suggest that azithromycin's blood-brain penetration may be better than doxycycline [37, 38]. Patients who cannot take adequate water or are not able to sit upright are at risk of pill esophagitis with doxycycline [39]. In such cases, azithromycin may be preferred.

In three studies of mild scrub typhus, ≤ 3 days of azithromycin was at least as good as 5-7 days of doxycycline. Patients may prefer the shorter duration and once-daily dosing of azithromycin. In theory, improved adherence to a shorter azithromycin regimen will prevent resistance. Limiting exposure to doxycycline in mild cases might theoretically prevent resistance to doxycycline in other co-existent pathogens such as malaria. A possible approach can be limiting the use of doxycycline to severe cases where there is some evidence supporting its efficacy for scrub typhus. In certain regions where the availability of intravenous doxycycline is limited, advocating for improved access might be beneficial.

It's essential to acknowledge the limitations of this study. Most included studies had a high risk of bias, which could affect the reliability of the findings. Heterogeneity was also observed in the meta-analysis of time to defervescence, possibly due to variations in study populations, dosages, and definitions of outcomes across studies. The doxycycline arm in one of the included RCTs also enrolled patients who received chloramphenicol, which decreased the overall reliability of the doxycycline vs. azithromycin comparison [25]. Most studies did not report mean and standard deviation for time to defervescence. They were calculated using indirect methods from the median, indicating that the primary data might have been skewed from the beginning [9].

Conclusions

The current data from studies with a high risk of bias did not find statistically significant differences in clinical outcomes between doxycycline and azithromycin for scrub typhus. The choice between these antibiotics should be considered based on availability, cost and patient profile. Further research, including well-designed randomised controlled trials with low risk of bias, is needed to provide more definitive evidence for guiding treatment decisions in scrub typhus.

Availability of data and materials

All data generated or analysed during this study are included in this published article [and its supplementary information files].

Abbreviations

AUFI:

Acute undifferentiated febrile illness

RCT:

Randomized controlled trial

nRCT:

Non-randomized clinical trials

RR:

Risk ratio

CI:

Confidence Interval

References

  1. Abhilash KPP, Jeevan JA, Mitra S, Paul N, Murugan TP, Rangaraj A, et al. Acute undifferentiated febrile illness in patients presenting to a tertiary care hospital in south india: clinical spectrum and outcome. J Glob Infect Dis. 2016;8(4):147–54.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Bonell A, Lubell Y, Newton PN, Crump JA, Paris DH. Estimating the burden of scrub typhus: a systematic review. PLoS Negl Trop Dis. 2017;11(9):e0005838.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Jiang J, Richards AL. Scrub typhus: no longer restricted to the tsutsugamushi triangle. Trop Med Infect Dis. 2018;3(1):11.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Saraswati K, Maguire BJ, McLean ARD, Singh-Phulgenda S, Ngu RC, Newton PN, et al. Systematic review of the scrub typhus treatment landscape: assessing the feasibility of an individual participant-level data (IPD) platform. PLoS Negl Trop Dis. 2021;15(10):e0009858.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Devasagayam E, Dayanand D, Kundu D, Kamath MS, Kirubakaran R, Varghese GM. The burden of scrub typhus in India: a systematic review. PLoS Negl Trop Dis. 2021;15(7):e0009619.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Rahi M, Gupte MD, Bhargava A, Varghese GM, Arora R. DHR-ICMR guidelines for diagnosis & management of rickettsial diseases in India. Indian J Med Res. 2015;141(4):417–22.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Higgins JPT, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;18(343):d5928.

    Article  Google Scholar 

  8. Sterne JA, Hernán MA, Reeves BC, Savović J, Berkman ND, Viswanathan M, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;12(355):i4919.

    Article  Google Scholar 

  9. Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol. 2014;14(1):135.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Song JH, Lee C, Chang WH, Choi SW, Choi JE, Kim YS, et al. Short-course doxycycline treatment versus conventional tetracycline therapy for scrub typhus: a multicenter randomized trial. Clin Infect Dis Off Publ Infect Dis Soc Am. 1995;21(3):506–10.

    Article  CAS  Google Scholar 

  11. Watt G, Kantipong P, Jongsakul K, Watcharapichat P, Phulsuksombati D, Strickman D. Doxycycline and rifampicin for mild scrub-typhus infections in northern Thailand: a randomised trial. Lancet Lond Engl. 2000;356(9235):1057–61.

    Article  CAS  Google Scholar 

  12. Silpapojakul K, Varachit B, Silpapojakul K. Paediatric scrub typhus in Thailand: a study of 73 confirmed cases. Trans R Soc Trop Med Hyg. 2004;98(6):354–9.

    Article  PubMed  Google Scholar 

  13. Zhao M, Wang T, Yuan X, Du W, Lin M, Shen Y. Comparison of minocycline and azithromycin for the treatment of mild scrub typhus in northern China. Int J Antimicrob Agents. 2016;48(3):317–20.

    Article  PubMed  CAS  Google Scholar 

  14. Lee M, Kim J, Jo DS. Effects of clarithromycin treatment in scrub typhus in children: comparison with chloramphenicol and azithromycin. Korean J Pediatr. 2017;60(4):124–7.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  15. Giri PP, Roy J, Saha A. Scrub typhus - a major cause of pediatric intensive care admission and multiple organ dysfunction syndrome: a single-center experience from India. Indian J Crit Care Med Peer-Rev Off Publ Indian Soc Crit Care Med. 2018;22(2):107–10.

    Google Scholar 

  16. Basu S, Saha A, Sarkar S, Sinha MK, Das MK, Datta R, et al. Clinical profile and therapeutic response of scrub typhus in children: a recent trend from Eastern India. J Trop Pediatr. 2019;65(2):139–46.

    Article  PubMed  Google Scholar 

  17. Ghosh A, Dey PK, Hemram SK, Mukherjee M, Nair A. Clinical profile and predictors of therapeutic outcome with azithromycin in pediatric scrub typhus: a study from Eastern India. Trop Doct. 2022;52(3):400–7.

    Article  PubMed  Google Scholar 

  18. Varghese GM, Dayanand D, Gunasekaran K, Kundu D, Wyawahare M, Sharma N, et al. Intravenous doxycycline, azithromycin, or both for severe scrub typhus. N Engl J Med. 2023;388(9):792–803.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  19. Guan XG, Zhou SX, Zhang AR, Lu QB, Zhou ZW, Chen JJ, et al. Clinical outcomes of doxycycline, azithromycin and chloramphenicol for the treatment of patients with severe scrub typhus. Int J Antimicrob Agents. 2022;60(4):106647.

    Article  PubMed  CAS  Google Scholar 

  20. Hwang JH, Kim MJ, Im YJ, Moon SJ, Kim JH, Lee MG, et al. Treatment outcomes of oral doxycycline versus intravenous azithromycin in adults hospitalized with scrub typhus: A retrospective study using inverse probability treatment weighting (IPTW) propensity analysis. Travel Med Infect Dis. 2023;1(52):102525.

    Article  Google Scholar 

  21. Kabir KI, John J, Satapathy AK, Sahu S, Behera B, Padhy BM. Oral azithromycin versus doxycycline in the treatment of children with uncomplicated scrub typhus: a randomized controlled trial. Pediatr Infect Dis J. 2022;41(3):224–9.

    Article  PubMed  Google Scholar 

  22. T AB, Dk N, L J. Prospective Study to Assess the Response to Therapy and Its Predictors in Children with Scrub Typhus. J Trop Pediatr. 2021;67(5). Cited 2023 Apr 28. Available from: https://pubmed.ncbi.nlm.nih.gov/34613376/.

  23. Barnabas R, Abhilash K, Varghese GM, Shubanker M, Ramya I, Prakash J. Prospective study to assess the treatment modalities and fever defervescence in patients with scrub typhus from a tertiary care centre in South India. J Vector Borne Dis. 2021;58(1):33–8.

    Article  PubMed  CAS  Google Scholar 

  24. Veerappan I, Ramar R, Palanisamy S. Antibiotic response to pediatric scrub typhus in South India: is clinical failure to azithromycin to be worried? J Trop Pediatr. 2021;67(1):fmab013.

    Article  PubMed  Google Scholar 

  25. Chanta C, Phloenchaiwanit P. Randomized controlled trial of azithromycin versus doxycycline or chloramphenicol for treatment of uncomplicated pediatric scrub typhus. J Med Assoc Thail Chotmaihet Thangphaet. 2015;98(8):756–60.

    Google Scholar 

  26. Phimda K, Hoontrakul S, Suttinont C, Chareonwat S, Losuwanaluk K, Chueasuwanchai S, et al. Doxycycline versus azithromycin for treatment of leptospirosis and scrub typhus. Antimicrob Agents Chemother. 2007;51(9):3259–63.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  27. Kim YS, Yun HJ, Shim SK, Koo SH, Kim SY, Kim S. A comparative trial of a single dose of azithromycin versus doxycycline for the treatment of mild scrub typhus. Clin Infect Dis. 2004;39(9):1329–35.

    Article  PubMed  CAS  Google Scholar 

  28. Yang J, Luo L, Chen T, Li L, Xu X, Zhang Y, et al. Efficacy and safety of antibiotics for treatment of scrub typhus. JAMA Netw Open. 2020;3(8):e2014487.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Lu D, Wang T, Luo Z, Ye F, Qian J, Zhang J, et al. Evaluation of the therapeutic effect of antibiotics on scrub typhus: a systematic review and network meta-analysis. Front Public Health. 2022;10:883945.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Gaillard T, Briolant S, Madamet M, Pradines B. The end of a dogma: the safety of doxycycline use in young children for malaria treatment. Malar J. 2017;13(16):148.

    Article  Google Scholar 

  31. Sharland M, Cappello B, Ombajo LA, Bazira J, Chitatanga R, Chuki P, et al. The WHO AWaRe Antibiotic Book: Providing guidance on optimal use and informing policy. Lancet Infect Dis. 2022;22(11):1528–30.

    Article  PubMed  Google Scholar 

  32. Gupta N, Joylin S, Ravindra P, Wilson W, Bhat R, Guddattu V, et al. A diagnostic randomised controlled trial to study the impact of rapid diagnostic tests in patients with Acute febrile illness when compared to conventional diagnostics (DRACO study). J Infect. 2021;82(6):e6-8.

    Article  PubMed  Google Scholar 

  33. Rajan SJ, Sathyendra S, Mathuram AJ. Scrub typhus in pregnancy: Maternal and fetal outcomes. Obstet Med. 2016;9(4):164–6. https://doi.org/10.1177/1753495X16638952.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Chung MH, Lee JS, Im JH. Antibiotic combination therapy for severe scrub typhus: is it necessary? Infect Chemother. 2023;55(2):179–84.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Mortensen EM, Halm EA, Pugh MJ, et al. Association of azithromycin with mortality and cardiovascular events among older patients hospitalized with pneumonia. JAMA. 2014;311(21):2199–208. https://doi.org/10.1001/jama.2014.4304.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  36. John J, Le T, Harrington N. Impact of doxycycline in place of azithromycin for community-acquired pneumonia on Clostridium difficile infections. Open Forum Infect Dis. 2018;5(Suppl 1):S191. https://doi.org/10.1093/ofid/ofy210.526.

    Article  PubMed Central  Google Scholar 

  37. Yim CW, Flynn NM, Fitzgerald FT. Penetration of oral doxycycline into the cerebrospinal fluid of patients with latent or neurosyphilis Antimicrob. Agents Chemother. 1985;28:347–8.

    Article  CAS  Google Scholar 

  38. Jaruratanasirikul S, Hortiwakul R, Tantisarasart T, Phuenpathom N, Tussanasunthornwong S. Distribution of azithromycin into brain tissue, cerebrospinal fluid, and aqueous humor of the eye Antimicrob. Agents Chemother. 1996;40:825–6.

    Article  CAS  Google Scholar 

  39. Kadayifci A, Gulsen MT, Koruk M, Savas MC. Doxycycline-induced pill esophagitis. Dis Esophagus. 2004;17(2):168–71.

    Article  PubMed  CAS  Google Scholar 

Download references

Acknowledgements

None.

Funding

No funding was received for the study.

Author information

Authors and Affiliations

  1. Department of Infectious Diseases, Kasturba Medical College, Manipal, Manipal Academy of Higher Education, Manipal, 576104, India

    Nitin Gupta

  2. Department of Clinical Sciences, Institute of Tropical Medicine, Nationalestraat 155, 2000, Antwerp, Belgium

    Nitin Gupta, Carl Boodman, Christelle Genevieve Jouego & Steven Van Den Broucke

  3. Division of Infectious Diseases, Department of Internal Medicine, University of Manitoba, Winnipeg, MB, Canada

    Carl Boodman

  4. Molecular Diagnostic and Research Group, University of Yaoundé, 11864, Yaoundé, Cameroon

    Christelle Genevieve Jouego

Authors
  1. Nitin Gupta
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Carl Boodman
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Christelle Genevieve Jouego
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Steven Van Den Broucke
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

The study was conceived and designed by NG and SVDB. The selection of articles, screening and data extraction was done by NG, CB and SVDB. The data analysis was done by NG and CGJ. All authors were involved in writing and revising the article.

Corresponding author

Correspondence to Steven Van Den Broucke.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1:

 Supplementary Table 1. Databases searched and search string used for the systematic review. Supplementary Table 2. Inclusion and Exclusion criteria used for screening and full-text review. Supplementary Table 3. Summary of studies with data available for either doxycycline or azithromycin but not both. Supplementary Table 4. Additional details of patient selection criteria in terms of the use of diagnostics, and previous antimicrobial use. Supplementary Table 5. Fever, mortality, and adverse events related to primary and secondary outcomes in the included studies. Supplementary Figure 1. Meta-analysis of doxycycline vs. azithromycin for time to defervescence classified according to age group. Supplementary Figure 2. Meta-analysis to calculate the standardised mean difference of time to defervescence between doxycycline and azithromycin with studies stratified according to the severity.Supplementary Figure 3. Mean difference of time to defervescence between doxycycline and azithromycin categorised according to whether loading dose was given or not. Supplementary Figure 4. Meta-analysis of doxycycline vs azithromycin showing the proportion of patients not achieving defervescence within five days of initiation of drugs. Supplementary Figure 5. Meta-analysis of doxycycline vs. azithromycin showing the proportion of patients with treatment-related adverse effects.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gupta, N., Boodman, C., Jouego, C.G. et al. Doxycycline vs azithromycin in patients with scrub typhus: a systematic review of literature and meta-analysis. BMC Infect Dis 23, 884 (2023). https://doi.org/10.1186/s12879-023-08893-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12879-023-08893-7

Keywords

BMC Infectious Diseases

ISSN: 1471-2334

Contact us