If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Internal Medicine, E-DA Cancer Hospital, Kaohsiung, TaiwanInstitute of Biomedical Sciences, National Sun Yat-sen University, Kaohsiung, Taiwan
Institute of Biomedical Sciences, National Sun Yat-sen University, Kaohsiung, TaiwanDepartment of Internal Medicine, E-DA Dachang Hospital, Kaohsiung, Taiwan
Division of General Surgery, Department of Surgery, E-Da Cancer Hospital, Kaohsiung, TaiwanSchool of Medicine, College of Medicine, I-Shou University, Kaohsiung, Taiwan
Department of Psychology, College of Medical and Health Science, Asia University, Taichung, TaiwanGender Equality Education and Research Center, Asia University, Taichung, TaiwanDepartment of Medical Research, Asia University Hospital, Asia University, Taichung, TaiwanDepartment of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan
⁎⁎⁎⁎ these authors contributed equally as corresponding authors.
Affiliations
Institute of Biomedical Sciences, National Sun Yat-sen University, Kaohsiung, TaiwanProspect Clinic for Otorhinolaryngology & Neurology, Kaohsiung, TaiwanDepartment of Psychology, College of Medical and Health Science, Asia University, Taichung, Taiwan
⁎⁎⁎⁎ these authors contributed equally as corresponding authors.
Chang-Hua Chen
Correspondence
Chang-Hua Chen, Department of Internal Medicine, Changhua Christian Hospital, Changhua, Taiwan, No. 135, Nanxiao Street, Changhua City, Changhua County 500, Taiwan, Tel: +886-4-7238595.
⁎⁎⁎⁎ these authors contributed equally as corresponding authors.
Affiliations
Program in Translational Medicine, National Chung Hsing University, Taichung City, TaiwanRong Hsing Research Center For Translational Medicine, National Chung Hsing University, Taichung City, TaiwanDepartment of Internal Medicine, Changhua Christian Hospital, Changhua, TaiwanPh.D. Program in Microbial Genomics, National Chung Hsing University and Academia Sinica, TaiwanDepartment of Post-Baccalaureate Medicine, College of Medicine, National Chung Hsing University, Taiwan
All the investigated antibiotics had similar cure rates to doxycycline.
•
The second-generation quinolone had lower adverse event rates than doxycycline.
•
Chloramphenicol and minocycline had the best cure rate in the child group and adult group, respectively.
Abstract
Objectives
Inadequate treatment of scrub typhus results in severe complications such as septic shock and is also associated with a high median mortality rate of 6%. However, there has been no conclusive evidence about the superiority of different antibiotics in managing scrub typhus in terms of efficacy and tolerability.
Methods
We conducted a network meta-analysis (NMA) using the frequentist method. The included participants were pediatric and adult patients infected with scrub typhus. The primary outcome was the cure rate in the patients included. The subgroup analysis was done according to pediatric or adult patients.
Results
Overall, 14 randomized controlled trials (RCTs) with 1264 participants were included in this study. The NMA revealed that all the investigated antibiotics were associated with cure rates similar to those of doxycycline. The chloramphenicol and minocycline were ranked to be associated with the highest cure rate in the pediatric subgroup and adult subgroup, respectively. Second-generation quinolones, including ofloxacin, ciprofloxacin, and chloramphenicol, were associated with significantly lower adverse event rates than doxycycline.
Conclusion
The current updated NMA provides evidence for the efficacy of chloramphenicol and minocycline in scrub typhus management. However, future large-scale RCTs with longer follow-up times are warranted.
). Furthermore, inadequate treatment of scrub typhus because of lack of awareness would result in severe complications, such as acute respiratory distress syndrome, acute renal failure, myocarditis, and septic shock (
Despite having a relatively high risk of mortality, there is little literature regarding treatments for managing scrub typhus. Doxycycline and azithromycin are the two most frequently recommended drugs for the treatment of scrub typhus (
). In the recent guidelines, doxycycline has been recommended as the drug of choice that can be prescribed without waiting for confirmatory laboratory results (
). In addition, tetracyclines are not recommended for pregnant women or children owing to its side effects. The occurrence of the gray baby syndrome in premature infants, a hematologic side effect of chloramphenicol, has also significantly reduced the use of chloramphenicol. Therefore, finding an effective regimen to treat scrub typhus is an urgent requirement that will also help control similar Rickettsial diseases.
Recently, there has been a network meta-analysis (NMA) published to address the efficacy and tolerability of different antibiotics to scrub typhus, which recommended that the clarithromycin might be the best antibiotic for alleviating fever in people with scrub typhus (
). However, in that NMA, the included studies had huge heterogeneity. Specifically, it consisted of both randomized controlled trials (RCTs) and case-control studies simultaneously, which disrupted the fundamental transitivity assumption in the context of NMA (
). However, according to the original statement in the article by Paul et al., they said “…Medications were given under supervision. With the exception of three patients who needed intravenous chloramphenicol because of the severity of their illness (one with meningoencephalitis, 2 others very toxic and obtunded), 37 patients were given oral tetracycline…” (
). Therefore, in this article by Paul et al., the assignment of antibiotics was based on the consideration of severity but not based on simply fair randomization, which might be resulted from the poor allocation concealment. The fundamental assumption of transitivity meant that the comparison between treatment A and B could be calculated from the comparison of A versus C and C versus B. However, this transitivity assumption was based on the faithful randomization process so that there is no imbalance in the effect modifiers across the different types of treatment comparisons (
). Therefore, to include non-RCTs in the NMA would disrupt the fundamental theoretical assumption of the NMA design so that the overall bias within the result of NMA would increase. In addition, in that previous NMA, the authors did not conduct subgroup analysis according to adult or pediatric patients. In the previous epidemiologic report, the infection of scrub typhus in adult patients was associated with more frequent and higher severity of complications (for example, shock, acute renal failure, and hepatic failure) than those in children (
). It would be of clinical importance to provide subgroup analysis based on adult or pediatric patients. Further, among the included studies of that NMA, some studies also recruited other infectious diseases, such as leptospirosis and murine typhus (
), so the inclusion of other infectious diseases might interfere with the result of the overall NMA. Therefore, the clinical application of the clarithromycin recommendation might remain controversial.
NMA of existing RCTs enables the comparison of the efficacy and the relative merits of multiple interventions. Moreover, it allows maximizing statistical power, which cannot be done in traditional pairwise meta-analysis (
). Considering these merits, we conducted an updated NMA of the published RCTs by estimating the relative efficacy and tolerability of different antibiotics in patients with scrub typhus.
Methods
The detailed description of the current NMA is depicted in Appendix: eMethods and materials. The overall structure of this NMA followed our previous NMA studies (
Efficacy and acceptability of different anti-fungal interventions in oropharyngeal or esophageal candidiasis in HIV co-infected adults: a pilot network meta-analysis.
). In brief, following the preferred reporting items for systematic reviews and meta-analyses 2020 guideline (eTable 1), the current NMA was conducted to include only published human RCTs, which investigated the efficacy of antibiotics in the treatment of scrub typhus (detailed search strategy is listed in eTable 2). The population, intervention, comparison, and outcome setting of the current NMA was as follows: (1) P: patients with an infection of scrub typhus, which could be supported by clinical presentation or serology test; (2) I: treatment with antibiotics; (3) C: the comparative interventions were compared with another antibiotics (to be specific, it would be unethical to provide a placebo to subjects with an active infectious episode); and (4) O: subsided target symptoms (i.e., fever). The primary outcome was the cure rate of the scrub typhus infection. The secondary outcomes were the fever duration after initiation of antibiotic treatment and the relapse rate. The tolerability was investigated according to the rate of any adverse event reported. For continuous variables, we estimated the effect size by evaluating the standardized mean difference (SMD) with 95% confidence intervals (95% CIs). For categorical variables, rate ratios (RRs) and 95% CIs (i.e., the cure rate or acceptability) were used, and a 0.5 zero-cell correction was applied during the meta-analysis procedure. We used the frequentist model of NMA to compare the effect size of studies with similar interventions. In the presentation of the forest plot, we chose doxycycline as the reference group because doxycycline had been recommended as the first line antibiotic to treat scrub typhus in patients of any age. We used mixed comparison with generalized linear mixed models to perform direct and indirect comparisons. STATA version 16.0 (StataCorp LLC Statistics/Data Analysis StataCorp, Texas, USA) was used in our NMA with the mvmeta command. To provide additional information for clinical applications, we calculated the surface under the cumulative ranking curve (SUCRA) to rank the probabilities of the effects of all treatments for target outcomes. To provide a more comprehensive clinical application, we arranged a subgroup analysis based on adult or pediatric subjects.
Results
In total, 42 articles were considered for full-text review (Figure 1), of which 28 were excluded according to the exclusion criteria (eTable 3). Among the excluded studies, the study by Paul et al. assigned the antibiotics partially based on the consideration of severity but not based on simply fair randomization, which might be resulted from the poor allocation concealment (
). Regarding the study by Chanta and Phloenchaiwanit, the authors initially classified the patients randomly into azithromycin versus doxycycline or chloramphenicol (
). However, in the later stage, the allocation to doxycycline or chloramphenicol was determined according to the age stratification but not simply fair randomization, which might be resulted from the poor allocation concealment. Therefore, it would be less feasible to regard this study as simply “azithromycin versus doxycycline versus chloramphenicol.” As previously mentioned, the fair randomization was the cornerstone of the transitivity hypothesis (
). Although the leptospirosis might have overlapped antibiotics susceptibility with that of scrub typhus, the inclusion of co-infection might potentially increase the heterogeneity in the current NMA through the statistical view. Therefore, this study was also excluded accordingly. Finally, 14 articles were included in the current study (eTable 4).
Figure 1Flowchart of the current network meta-analysis.Figure 1.Figure 1 depicts the entire flowchart of the current network meta-analysis.
A total of 1264 participants were included. The mean age of the participants was 45.2 years (range 5.5-83.0 years), and the mean female proportion was 48.8% (range 8.3-70.2%). The mean treatment duration was 6.4 days (range 5-10 days). The baseline characteristics of the included participants are listed in eTable 4. Among the included RCTs, nine recruited adult patients, whereas three only included pediatric patients. Because of the ethical consideration, none of the included RCTs applied placebo-controlled treatment.
The definition of cure rate applied in the included RCTs included the following: (1) resolution of fever within 3-5 days after initiating antimicrobial therapy, (2) resolution of fever without relapse in one week after discontinuing the antimicrobial therapy, and (3) complete resolution of fever and other symptoms as well as the moderation of laboratory abnormalities at the end of therapy. The definition of relapse rate included the following: (1) reappearance of fever and clinical manifestations of scrub typhus within 30 days after completing therapy, (2) fever or other clinical symptoms recurring within 30 days after determining the patient as completely cured, and (3) recurrence of fever and other symptoms during the follow-up period (four weeks).
Regarding the treatment arm definition, in the study by Quan, we kept the strategy of pooling ofloxacin and ciprofloxacin into “quinolone” because of the rationale mentioned later (
). First, both ofloxacin and ciprofloxacin share a similar chemical structure and inhibit the same enzymatic activities of DNA gyrase and topoisomerase IV to exert their antibiotic effect (
). In addition, the major issue of the quinolone resistance in scrub typhus management resulted from the gyrA mutation, which will affect the efficacy of both ofloxacin and ciprofloxacin (
). Therefore, it would be a reasonable rationale to consider the quinolone as one entire category.
In contrast, although tetracycline and minocycline have the same properties in general, they exert different efficacy in clinical practice. For example, minocycline has a different structure from that of tetracycline and was developed initially to manage a large number of tetracycline-resistant microbiomes, which are termed the tetracycline resistome (
). In addition, tetracycline is a short-acting formulation, but minocycline is a long-acting formulation. Furthermore, tetracycline and minocycline have different safety profiles. Specifically, an adverse reaction of tetracycline is tooth discoloration, but that of minocycline is dizziness (9%), headache (23%), and fatigue (9%). Therefore, it would be rational to classify tetracycline and minocycline into different treatment arms.
Primary outcome: cure rate
The NMA revealed that all the investigated antibiotics were associated with similar cure rates compared with doxycycline (Table 1A; Figure 2A and 3A). According to the SUCRA, minocycline (RR = 1.05, 95% CIs = 0.96-1.14 compared with doxycycline) was associated with the highest cure rate, followed by tetracycline (RR = 1.03, 95% CIs = 0.97-1.11 compared with doxycycline), and telithromycin (RR = 1.02, 95% CIs = 0.96-1.09 compared with doxycycline) (eTable 5).
Table 1ALeague table of the cure rate
Min
1.03 (0.96,1.12)
1.01 (0.91,1.13)
Ttr
*0.62 (0.39,0.99)
1.00 (0.84,1.19)
1.02 (0.92,1.14)
1.01 (0.92,1.11)
Tel
1.02 (0.96,1.09)
1.03 (0.95,1.12)
1.02 (0.95,1.10)
1.01 (0.94,1.08)
Azi
1.00 (0.93,1.08)
1.00 (0.97,1.03)
1.00 (0.93,1.08)
1.02 (0.98,1.06)
1.03 (0.94,1.14)
1.02 (0.94,1.11)
1.01 (0.93,1.10)
1.00 (0.95,1.06)
Qui
1.01 (0.95,1.07)
1.00 (0.93,1.08)
1.00 (0.93,1.07)
1.04 (0.95,1.13)
1.02 (0.95,1.11)
1.01 (0.94,1.09)
1.00 (0.97,1.03)
1.00 (0.95,1.06)
Chl
1.00 (0.93,1.08)
0.96 (0.84,1.09)
1.05 (0.96,1.14)
1.03 (0.96,1.11)
1.02 (0.96,1.09)
1.01 (0.97,1.05)
1.01 (0.95,1.07)
1.01 (0.97,1.05)
Rif
1.00 (0.98,1.02)
1.05 (0.96,1.14)
1.03 (0.97,1.11)
1.02 (0.96,1.09)
1.01 (0.98,1.05)
1.01 (0.96,1.07)
1.01 (0.97,1.05)
1.00 (0.98,1.02)
Dox
Azi = azithromycin; Chl = chloramphenicol; Dox = doxycycline; Min = minocycline; Qui = second-generation quinolone; Rif = rifampicin; Tel = telithromycin; Ttr = tetracycline.
Pairwise (upper-right portion) and network (lower-left portion) meta-analysis results are presented as estimate effect sizes for the outcome of the cure rate in (A) overall patients with scrub typhus, (B) subgroup of pediatric patients, and (C) subgroup of adult patients. Interventions are reported in the order of mean rank of efficacy, and outcomes are expressed as rate ratios (RRs) (95% confidence interval). For the pairwise meta-analyses, an RR of more than 1 indicates that the treatment specified in the row had a better cure rate than that specified in the column. For the network meta-analysis (NMA), an RR of more than 1 indicates that the treatment specified in the column had a better cure rate than that specified in the row. Bold results marked with * indicate statistical significance.
Figure 2The network structure of (2A) cure rates, (2B) cure rates in subgroup of pediatric patients, and (2C) cure rates in subgroup of adult patients.Figure 2 depicts the overall network structure of the current network meta-analysis of the (2A) cure rates, (2B) cure rates in subgroup of pediatric patients, and (2C) cure rates in subgroup of adult patients. The lines between nodes represent direct comparisons in various trials, and the size of each circle is proportional to the size of the population involved in each specific treatment. The thickness of the lines is proportional to the number of trials connected to the network.
Figure 3Forest plot of the (3A) cure rates, (3B) cure rates in subgroup of pediatric patients, and (3C) cure rates in subgroup of adult patients.Figure 3 indicates that when the effect size is more than one, the specified treatment is associated with higher cure rate in patients with scrub typhus than doxycycline in (3A) overall subjects, (3B) in subgroup of pediatric patients, and (3C) in subgroup of adult patients.
The NMA revealed that all the investigated antibiotics were associated with similar cure rates compared with doxycycline (Table 1B; Figure 2B and 3B). According to the SUCRA, chloramphenicol was associated with the highest cure rate (RR = 1.02, 95% CIs = 0.95-1.09 compared with doxycycline) (eTable 5).
Table 1BLeague table of the cure rate: subgroup of pediatric patients
Chl
1.00 (0.97,1.04)
1.00 (0.97,1.04)
Azi
1.02 (0.96,1.08)
1.02 (0.95,1.09)
1.02 (0.96,1.08)
Dox
Abbreviations: Azi = azithromycin; Chl = chloramphenicol; Dox = doxycycline.
The NMA revealed that all the investigated antibiotics were associated with similar cure rates compared with doxycycline (Table 1C; Figure 2C and 3C). According to the SUCRA, minocycline was associated with the highest cure rate (RR = 1.06, 95% CIs = 0.96-1.16 compared with doxycycline) (eTable 5).
Table 1CLeague table of the cure rate: subgroup of adult patients
Min
1.03 (0.96,1.12)
1.02 (0.91,1.14)
Ttr
*0.62 (0.39,0.99)
1.00 (0.84,1.19)
1.03 (0.95,1.12)
1.01 (0.93,1.10)
Azi
1.00 (0.93,1.07)
1.00 (0.93,1.08)
1.00 (0.93,1.08)
1.03 (0.96,1.10)
1.03 (0.93,1.15)
1.01 (0.92,1.11)
1.00 (0.93,1.08)
Tel
1.02 (0.96,1.09)
1.04 (0.94,1.15)
1.02 (0.93,1.11)
1.01 (0.95,1.08)
1.01 (0.93,1.09)
Chl
1.00 (0.93,1.08)
1.00 (0.93,1.08)
1.00 (0.94,1.07)
1.05 (0.94,1.16)
1.02 (0.94,1.12)
1.01 (0.95,1.09)
1.01 (0.93,1.10)
1.01 (0.94,1.08)
Qui
1.00 (0.93,1.08)
1.00 (0.93,1.07)
1.06 (0.96,1.16)
1.03 (0.96,1.11)
1.02 (0.97,1.08)
1.02 (0.96,1.09)
1.01 (0.96,1.08)
1.01 (0.95,1.07)
Rif
1.00 (0.98,1.02)
1.06 (0.96,1.16)
1.03 (0.97,1.11)
1.02 (0.98,1.08)
1.02 (0.96,1.09)
1.02 (0.96,1.08)
1.01 (0.95,1.07)
1.00 (0.98,1.02)
Dox
Pairwise (upper-right portion) and network (lower-left portion) meta-analysis results are presented as estimate effect sizes for the outcome of the cure rate in (A) overall patients with scrub typhus, (B) subgroup of pediatric patients, and (C) subgroup of adult patients.
Interventions are reported in the order of mean rank of efficacy, and outcomes are expressed as rate ratios (RRs) (95% confidence interval). For the pairwise meta-analyses, an RR of more than 1 indicates that the treatment specified in the row had a better cure rate than that specified in the column. For the network meta-analysis (NMA), an RR of more than 1 indicates that the treatment specified in the column had a better cure rate than that specified in the row. Bold results marked with *indicate statistical significance.
Secondary outcome: fever duration after initiation of antibiotics treatment
The NMA revealed that all the investigated antibiotics were associated with similar fever duration after initiation of antibiotic treatment compared with doxycycline in the participants (eTable 6A; eFigure 1A and 2A). According to the SUCRA, chloramphenicol was associated with the shortest fever duration after initiation of treatment (SMD = -0.76, 95% CIs = -2.08-0.56 compared with doxycycline) (eTable 5).
Secondary outcome: relapse rate
The NMA revealed that all the investigated antibiotics were associated with similar relapse rates compared with doxycycline in the study participants (eTable 6B; eFigure 1B and 2B). According to the SUCRA, the telithromycin was associated with the lowest relapse rate (RR = 0.32, 95% CIs = 0.01-7.64 compared with doxycycline) (eTable 5).
Tolerability reflected by adverse event rate
In the NMA, second-generation quinolones (RR = 0.18, 95% CIs = 0.06-0.50) and chloramphenicol (RR = 0.33, 95% CIs = 0.14-0.82) were associated with significantly lower adverse event rate than doxycycline. In contrast, tetracycline was associated with significantly higher adverse event rate than doxycycline (RR = 2.54, 95% CIs = 1.45-4.45) (eTable 6C; eFigure 1C and 2C). According to the SUCRA, second-generation quinolones were associated with the lowest adverse event rate (eTable 5).
Risk of bias, publication bias, inconsistency assessment, and Grading of Recommendations Assessment, Development and Evaluation (GRADE) ratings
We found that 44.9% (44/98 items), 21.4% (21/98 items), and 33.7% (33/98 items) of the included studies had a low, unclear, and high risk of bias, respectively. Unclear reporting of the allocation procedures and participant or research personnel blinding was the most common reason for the high risk of bias (eFigure 3A, B). Funnel plots of publication bias (eFigure 4A-H) revealed general symmetry, and the results of Egger's test indicated no significant publication bias among the articles included in the NMA. In general, the NMA did not demonstrate inconsistencies regarding either local inconsistencies, as assessed using the loop-specific approach and node-splitting method, or global inconsistencies, as determined using the design-by-treatment method (eTable 7, 8). Therefore, this examination might support the rationale of the basic hypothesis of similarity and transitivity in the current NMA. The results of GRADE evaluation have been listed in the Appendix. In brief, the overall quality of evidence of the overall NMA, direct evidence, and indirect evidence were low to medium (eTable 9).
Discussion
The current NMA revealed that no single investigated antibiotic was associated with significantly different cure rates compared with doxycycline in the overall number of patients with scrub typhus. In the subgroup of pediatric patients, chloramphenicol was associated with the highest cure rate among all the investigated antibiotics. In contrast, minocycline was associated with the highest cure rate in the subgroup of adult patients. Further, the second-generation quinolone, including ofloxacin, ciprofloxacin, and chloramphenicol, were associated with significantly lower adverse event rates than doxycycline.
By applying several different strategies from the previous NMA (
), we intended to improve the overall quality of the current NMA, enhance the similarity and the transitivity, and reduce the heterogeneity. Specifically, we only included RCTs and excluded non-RCTs. The inclusion of non-RCTs would disrupt the fundamental transitivity assumption in the context of NMA so that the overall reliability of NMA would consequently decrease (
). In addition, we excluded those RCTs that also consisted of other infectious diseases. That is, we only included RCTs of scrub typhus infection. The strategy of focusing target (i.e., scrub typhus) would help to avoid unwanted confounding effects of the concomitant infection (
). Specifically, although the leptospirosis might have overlapped antibiotics susceptibility with that of scrub typhus, the inclusion of a co-infection might potentially increase the heterogeneity in the current NMA through the statistical view. Therefore, the inclusion of other infectious diseases might interfere with the result of the overall NMA. That was the reason why we needed an NMA focusing specifically on RCTs of scrub typhus only. Finally, because of the epidemiological difference between scrub typhus in adult and pediatric patients, we arranged a subgroup analysis based on adult and pediatric patients in advance.
One of the most important findings of the current NMA was that minocycline was among the most effective antibiotics among all the antibiotics investigated in adult patients with scrub typhus. As previously mentioned, the scrub typhus infection in adult patients was associated with significantly more frequent and severe complications than in pediatric patients (
). Previously, there were no meta-analyses that addressed the efficacy and tolerability of minocycline in patients with scrub typhus. Along with the occurrence of doxycycline-resistant scrub typhus (
An outbreak of scrub typhus in military personnel despite protocols for antibiotic prophylaxis: doxycycline resistance excluded by a quantitative PCR-based susceptibility assay.
), other new antibiotics for scrub typhus management should be warranted. As O. tsutsugamushi is an obligate intracellular pathogen, antibiotics with good cell membrane penetration ability would be a better choice than those without (
). Minocycline exhibited good cell membrane penetration ability and long tissue half-life compared with its original analog, doxycycline, which allowed minocycline to be a candidate in the treatment of scrub typhus (
). In the current NMA, although without achieving statistical significance, minocycline was associated with a relatively better cure rate, shorter fever duration, and less adverse event rate than doxycycline. However, because of only one RCT investigating the efficacy of minocycline in the current NMA, future large-scale RCTs should be conducted to support or refute the result of the current NMA.
Another important finding of the current NMA was that second-generation quinolones were associated with significantly lower adverse event rates than doxycycline and were ranked to be associated with the lowest adverse event rates among all the antibiotics investigated for scrub typhus. Second-generation quinolones, including ofloxacin and ciprofloxacin, are well-known for their broad-spectrum bactericidal activity, which is achieved by binding and blocking DNA-type IIA topoisomerase complexes of bacteria (
). To date, no meta-analyses have assessed the efficacy of second-generation quinolones in patients with scrub typhus. There are only two RCTs addressing the efficacy and tolerability of second-generation quinolones in patients with scrub typhus (
). Therefore, the results of the current NMA support the use of second-generation quinolones, including ofloxacin and ciprofloxacin, for the management of scrub typhus. However, further experiments are needed to elucidate the safety of these drugs because the US Food and Drug Administration, in its 2016 safety review, has advised that systemic fluoroquinolone use is associated with uncommon but potentially adverse effects (
Food and Drug Administration, 2016. Drug UFDA. Safety communication: FDA advises restricting fluoroquinolone antibiotic use for certain uncomplicated infections; warns about disabling side effects that can occur together. http://www.fda.gov/Drugs/DrugSafety/ucm500143.htm (accessed on 30 March 2022).
There are several limitations to be addressed in the current NMA. First, some analyses in this study were limited by underpowered statistics, including heterogeneity in the characteristics of the participants (e.g., wide range of ages, lack of uniform criteria for definition of cure and relapse, and wide variety of trial duration and antibiotics dosage), and the small number of trials for some treatment arms. Second, we did not exclude trials with zero cases in both the intervention and control arms of one study because most RCTs had relapse cases equal to zero. This will risk increasing the bias (
). Third, although we tried to investigate the tolerability by analyzing the adverse event rate, none of the included RCTs reported any serious adverse events. This would contribute to the risk of omitting the potential rare but serious adverse events of specific antibiotics, such as intracranial hypertension and hypersensitivity reaction by minocycline (
French Network of Regional Centers of Pharmacovigilance. Comparative analysis of adverse drug reactions to tetracyclines: results of a French national survey and review of the literature.
). Fourth, we could not perform further subgroup analysis based on the different strains of O. tsutsugamushi or strains with specific antibiotic resistance. As mentioned in the introduction, doxycycline-resistant O. tsutsugamushi has been noticed in several parts of the Pacific Rim (
An outbreak of scrub typhus in military personnel despite protocols for antibiotic prophylaxis: doxycycline resistance excluded by a quantitative PCR-based susceptibility assay.
). From the current NMA, we could not determine whether doxycycline-resistant O. tsutsugamushi is resistant to other antibiotics with structures similar to that of doxycycline, such as minocycline and tetracycline. Fifth, we did not know whether the distribution of age or gender proportion of the included studies was normal distribution or not. Sixth, because of the relatively weak network structure, some comparisons lacked direct comparison (such as quinolone vs doxycycline in the rate of any adverse event). Finally, although our study is strengthened by comparing different antibiotics using NMA, the generalization of our results is still highly limited depending on the studies included and the possible comparisons within them. Future studies are warranted to assess the efficacy and tolerability of different antibiotics, focusing on the optimal dosage and treatment duration for managing scrub typhus caused by various strains of O. tsutsugamushi. Clinicians should consider specific strategies in specific clinical conditions.
The current updated NMA provided evidence for the efficacy and tolerability of individual antibiotics in scrub typhus management. In this study, we found that several investigated antibiotics were associated with a non-inferior cure rate compared with doxycycline in the subgroup of pediatric and adult patients. Furthermore, second-generation quinolones and chloramphenicol were associated with significantly lower adverse event rates than doxycycline. However, because of the small number of studies included, future large-scale RCTs focusing on the efficacy of different dosages and treatment duration of antibiotics in patients with scrub typhus should be conducted to support or refute the result of the current NMA.
Declaration of interests
The authors have no competing interests to declare. It was not appropriate or possible to involve patients or the public in the design, conduct, reporting, or dissemination plans of our research.
Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Ethics approval
The current study was approved by the Institutional Review Board of the Tri-Service General Hospital, National Defense Medical Center (TSGHIRB No. B-109–29).
Data sharing statement
All the data and materials of the current study are available upon reasonable request.
Acknowledgment
This paper presents independent research. The views expressed in this publication are those of the authors and not necessarily those of the acknowledged institutions.
Author contributions
Bing-Syuan Zeng, Bing-Yan Zeng, Chao-Ming Hung, and Hung-Chang Kuo, contributed equally as first authors and took full responsibility for literature screening, data extraction, data calculation, and manuscript drafting.
Yen-Wen Chen and Mein-Woei Suen took responsibility for concept formation, study design, literature selection and screening, supervision, visualization, data curation, and manuscript revision.
Yow-Ling Shiue, Ping-Tao Tseng, and Chang-Hua Chen contributed equally as corresponding authors and took full responsibility for software management, data calculation, data curation, concept formation, major manuscript revision, and submission.
Patient and public involvement statement
The current study did not directly involve individual participants, so we did not have the chance to approach individual participants or explore individual participants’ information. Therefore, it would be impossible to obtain the information on (1) consent to participate or publication, (2) how was the development of the research question and outcome measures informed by patients’ priorities, experience, and preferences, (3) how did we involve patients in the design of this study, (4) were patients involved in the recruitment to and conduct of the study, (5) how will the results be disseminated to study participants, (6) was the burden of the intervention assessed by patients themselves, and (7) patient advisers in the current study.
Role of the sponsors
None of the sponsors had been involved in the current study.
Food and Drug Administration, 2016. Drug UFDA. Safety communication: FDA advises restricting fluoroquinolone antibiotic use for certain uncomplicated infections; warns about disabling side effects that can occur together. http://www.fda.gov/Drugs/DrugSafety/ucm500143.htm (accessed on 30 March 2022).
An outbreak of scrub typhus in military personnel despite protocols for antibiotic prophylaxis: doxycycline resistance excluded by a quantitative PCR-based susceptibility assay.
French Network of Regional Centers of Pharmacovigilance. Comparative analysis of adverse drug reactions to tetracyclines: results of a French national survey and review of the literature.
Efficacy and acceptability of different anti-fungal interventions in oropharyngeal or esophageal candidiasis in HIV co-infected adults: a pilot network meta-analysis.
Trial registration: PROSPERO CRD42020222410. The current study was approved by the Institutional Review Board of the Tri-Service General Hospital, National Defense Medical Center (TSGHIRB No. B-109–29).
00358-7&id=gr1.jpg){kind=link}
00358-7&id=gr2.jpg){kind=link}
00358-7&id=gr3.jpg){kind=link}