Intrawound (local) antibiotic prophylaxis as an adjunct to systemic antibiotics to prevent periprosthetic joint infection (PJI) after total hip arthroplasty: a meta-analysis of cohort studies

Abstract

Objective

Surgical site infections (SSIs) particularly periprosthetic joint infection (PJI, a deep/organ-space infection) remain devastating after total hip arthroplasty (THA). Because routine systemic perioperative antibiotics are standard of care, we focused on the added value of intrawound (local) prophylaxis. We evaluated whether intrawound antibiotics, used in addition to standard systemic prophylaxis, reduce infection after THA and explored effects across surgery type (primary vs. revision).

Methods

We searched PubMed, Cochrane Library, ScienceDirect, EMBASE, CNKI, VIP, Wanfang, and CBM (January 2010 to final search) for cohort studies comparing intrawound antibiotic prophylaxis + standard systemic prophylaxis versus standard systemic prophylaxis alone. We systematically searched studies published since 2010 to include the most recent and relevant data, ensuring that the research included reflects modern surgical techniques, standardized reporting, and advancements in the use of intrawound antibiotics. Two reviewers independently screened, extracted data, and assessed risk of bias; meta-analysis used RevMan 5.3.

Results

Eight cohort studies (n = 16,939) met criteria. Heterogeneity for PJI was low (Chi² = 7.57, df = 7, P = 0.37; I² = 8%). Intrawound prophylaxis plus systemic prophylaxis was associated with lower PJI risk than systemic prophylaxis alone (OR = 0.42, 95% CI: 0.30–0.58, P < 0.00001). Subgroup analysis showed consistent benefit in both primary and revision THA (OR = 0.30, 95% CI: 0.20–0.47, P < 0.00001; I² = 0%). Superficial incisional SSI (skin/subcutaneous) showed no significant difference (OR = 0.41, 95% CI: 0.10–1.66, P = 0.92), and adverse events did not differ significantly (OR = 0.89, 95% CI: 0.45–1.78, P = 0.75). Funnel plots suggested possible publication bias.

Conclusion

There is limited evidence, with a possible risk of publication bias, suggesting that intrawound (local) antibiotic prophylaxis, when added to standard systemic antibiotics, may be associated with reduced postoperative PJI after THA across primary and revision procedures. Evidence for superficial SSI and safety remains limited by few contributing studies. Standardized, head-to-head randomized trials are needed to define optimal regimens (agent, dose, timing) and long-term safety (including antimicrobial resistance).

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-026-06722-6.

Introduction

Total hip arthroplasty (THA) is a common surgical method for treating end-stage hip joint diseases. It can significantly improve the joint function of patients, relieve pain, and enhance the quality of life [1–3]. Although the incidence of surgical site infection (SSI) has been decreasing with the improvement of joint replacement surgery techniques and perioperative preventive measures, the total number of SSIs has also been rising with the increasing number of surgical patients [4]. SSIs includes both superficial infections and periprosthetic joint infections (PJI) [5, 6]. Among these, PJI represents one of the most devastating complications after joint replacement surgery, often necessitating multiple revision procedures and prolonged antimicrobial therapy, thereby imposing a substantial economic and psychosocial burden on patients [7, 8].

The incidence of SSIs is influenced by multiple factors, including the patient’s advanced age, concurrent underlying diseases such as diabetes, large intraoperative blood loss, long operation time, long drainage time, and low preoperative serum albumin level, etc [9]. The administration of antibiotics is a significant factor influencing the incidence of infection. Studies have shown that prophylactic use of systemic antibiotics can reduce the incidence of infection after THA [10, 11] Current clinical guidelines recommend the rational use of antibiotics in the perioperative period to reduce the risk of SSIs. However, substantial variability exists in clinical practice regarding antibiotic regimens, including timing of administration, choice of agents, dosage, and duration of therapy. Inappropriate or prolonged antibiotic use may promote the emergence of drug-resistant pathogens, posing new challenges to infection prevention and control. While systemic antibiotic prophylaxis is a well-established practice with strong evidence supporting its efficacy, the role of additional adjuvant measures, such as intraound antibiotics, remains less clear, with some studies reporting significant benefits and others showing minimal effect intrawound antibiotics refer specifically to antibiotics applied directly into the wound during surgery (e.g., antibiotic powders or solutions), while intrawound antibiotics is a more specific term for antibiotics applied directly at the surgical site during the operation. Studies focusing on intrawound antibiotics (such as antibiotic powders or solutions) have yielded conflicting results, with varying protocols and a lack of uniformity in the studies. This variability limits the ability to conclusively determine their benefit in reducing PJI or superficial SSIs.

In addition to systemic antibiotics, other infection prevention strategies, such as Staphylococcus aureus decolonization, are also essential in reducing infection rates after THA. Several studies, including the systematic review and meta-analysis by Bianco Prevot et al. (2024), have demonstrated the effectiveness of Staphylococcus aureus decolonization as a mandatory procedure in preoperative knee and hip replacement care, significantly reducing periprosthetic infection rates. Nasal decolonization and other decolonization protocols play a key role in reducing the burden of infections in these high-risk patients [12]. Despite widespread use of perioperative systemic antibiotics, uncertainty persists regarding the added value of intrawound antibiotic prophylaxis in THA because prior studies report conflicting results and heterogeneous protocols. We therefore conducted a comprehensive meta-analysis of cohort studies to estimate the effect of intrawound prophylactic antibiotics on PJI and superficial SSI after primary and revision THA, and to appraise safety outcomes. By synthesizing evidence from multiple studies, this analysis seeks to furnish more comprehensive data to inform clinical decision-making, promote the rational use of antibiotics, reduce postoperative infection rates, and ultimately enhance therapeutic results and the quality of life for patients.

Methods

Protocol and reporting

This review followed PRISMA 2020 guidelines. The protocol was not prospectively registered; this is acknowledged as a study limitation.

Data sources and search strategy

We systematically searched PubMed, ScienceDirect, EMBASE, the Cochrane Library, and CNKI; supplementary sources included VIP, CBM, and Wanfang, covering Chinese and non-Chinese journals, conference proceedings, and dissertations. Reference lists of eligible studies were screened manually. Searches combined controlled vocabulary and free-text terms related to total hip arthroplasty, surgical site infection, periprosthetic joint infection, intrawound antibiotics (e.g., vancomycin powder), and prophylaxis. The time window spanned January 2010 through 23rd September 2025. Full database-specific search strings and exact search dates are provided in Supplementary Table S1. No language restrictions were applied; duplicates were removed prior to screening.

The decision to include studies published from 2010 onwards was made to ensure the inclusion of the most relevant and recent data. This period saw improvements in the quality of research, with more standardized reporting of outcomes, such as periprosthetic joint infection (PJI), superficial surgical site infection (SSI), and adverse events. Additionally, the years following 2010 saw a greater focus on intrawound antibiotics as an adjunct to systemic prophylaxis in hip arthroplasty, which aligned with the objectives of our study. Studies published before this period were excluded due to concerns about potential heterogeneity in study design, surgical protocols, and outcome definitions.

Eligibility criteria (PICOS)

Inclusion criteria

1. Research methodology: Cohort studies were included due to their relevance in real-world clinical settings, where randomized controlled trials (RCTs) are limited in the context of intrawound antibiotic prophylaxis.

2. Study population: Adults undergoing primary or revision total hip arthroplasty (THA).

3. Intervention: Intrawound antibiotic prophylaxis (e.g., antibiotic powder or solution applied into the wound during surgery) in addition to standard systemic perioperative prophylaxis.

4. Comparator: Standard systemic perioperative prophylaxis alone (i.e., no intrawound antibiotic).

5. Outcome measures: Primary periprosthetic joint infection (PJI); Secondary superficial SSI and adverse events.

6. Study design: Cohort studies were included due to their relevance in real-world clinical settings, where randomized controlled trials (RCTs) are limited in the context of intrawound antibiotic prophylaxis.

Exclusion criteria

1. Non-cohort study designs.

2. Incomplete data reporting or data unsuitable for analysis.

3. Duplicate publications, in which case the most recent or comprehensive version was included.

4. Unclear or unreported study outcomes.

5. Review articles.

6. Case reports.

Study selection

Two reviewers independently screened titles/abstracts and full texts, applying the criteria above. Disagreements were resolved by discussion or a third reviewer. Reasons for exclusion at full-text stage were recorded (PRISMA flow diagram).

This meta-analysis focused exclusively on studies that evaluated the use of intrawound antibiotics (e.g., antibiotic powders or solutions) applied directly to the surgical wound during surgery, in addition to standard systemic perioperative prophylaxis. The methods of application varied across studies and included techniques such as direct powder placement or antibiotic solution irrigation. Studies involving other forms of local antibiotic application, such as antibiotic-loaded cement or topical antimicrobial agents (e.g., hydrocolloid gels), were not included in the analysis.

Quality assessment and data extraction

Using a piloted form, two reviewers independently extracted: study identifiers; design; sample size; surgery type (primary/revision); definitions of PJI and superficial SSI; follow-up; and detailed intervention data (intrawound agent, dose, timing/application) for both groups, confirming standard systemic prophylaxis in all arms.Patient characteristics (age, gender, comorbidities such as diabetes, obesity, etc.) were also recorded to compare the groups receiving intrawound antibiotics vs. standard intravenous prophylaxis. Outcome events/participants were recorded for PJI, superficial SSI, and adverse events. Authors were contacted for missing data when needed.

Risk-of-bias assessment

Non-randomized studies were assessed with ROBINS-I across seven domains (confounding; selection; classification of interventions; deviations; missing data; outcome measurement; selection of reported results). Domain-level and overall judgments are shown in Figs. 2 and 3 and the Supplement.

Fig. 1.

Flowchart of literature screening

Fig. 2.

Risk bias map

Statistical analysis

Binary outcomes were pooled as odds ratios (OR) with 95% CIs (Mantel–Haenszel).Given the expected variability in study protocols (e.g., agent, dose, timing), a random-effects model was prespecified as the primary approach; a fixed-effect model was used as a sensitivity analysis when I² was low. Heterogeneity was quantified using the Q statistic and I². Heterogeneity was considered significant if I² > 50%, indicating moderate to high variability across studies. When I² > 50%, subgroup analyses and sensitivity checks were conducted to explore the potential sources of heterogeneity. In cases of significant heterogeneity, the random-effects model was maintained to account for between-study variability.

Prespecified subgroup analyses included primary vs. revision THA, while exploratory analyses (data permitting) investigated variations by agent, dose, and timing/application. Leave-one-out analyses were conducted to assess the influence of individual studies on the overall effect estimate. Subgroup analyses were conducted to evaluate the effects of intrawound antibiotics separately for primary THA and revision THA. Revision THA was further divided into septic and aseptic revisions to account for differences in infection risk and treatment protocols. Small-study effects/publication bias were evaluated using funnel plots and Egger’s test (α = 0.10); where asymmetry suggested bias, trim-and-fill was applied. Certainty of evidence for key outcomes was appraised using GRADE. Meta-analyses were performed in RevMan 5.3.

Results

Study selection and characteristics

Following PRISMA, 954 records were identified; 676 unique records remained after de-duplication. Title/abstract screening excluded 451 items; 325 full texts were assessed, and 317 were excluded (insufficient data or no prespecified outcomes). Eight cohort studies (n = 16,939) met criteria, all comparing intrawound/local prophylaxis + standard systemic antibiotics vs. standard systemic antibiotics alone in primary and/or revision THA. Figure 1 shows the PRISMA flow; Table 1 summarizes study characteristics.

Table 1.

Basic characteristics of included studies

Included studies	Year of publication	Sample size	Study design	Types of surgery	Intravenous preventive medication	Follow-up duration	Score for literature quality evaluation	Patient characteristics (age, gender, comorbidities)	Outcome Measures (PJI, SSI, AE)
Tian F [13]	2017	387	Retrospective study cohort	primary THA	Cefazolin 1 g before operation, 1 g after operation, Q8 hours, 48 h	Two years	4	Age: 65 ± 10, 70% Male, Diabetes (25%)	PJI: 3%, SSI: 1%, AE: 2%
Otte [14]	2017	1640	Retrospective study cohort	primary THA/ revision THA	Standard system prevention plan	Three months	6	Age: 70 ± 12, 60% Female, Obesity (30%)	PJI: 2.5%, SSI: 0.5%, AE: 1%
Dial [15]	2019	265	Retrospective study cohort	primary THA	Cefazolin 2 g was administered before the operation, and the drug was discontinued 24 h after the operation	≥ 3 months	4	Age: 68 ± 12, 55% Male, Hypertension (20%)	PJI: 1%, SSI: 0.7%, AE: 1.5%
Riesgo [16]	2018	74	Retrospective study cohort	revision of THA	The perioperative preventive medication for infectious microorganisms should be administered for at least 6 weeks	More than one year	5	Age: 70 ± 15, 50% Male, Obesity (15%)	PJI: 2%, SSI: 1%, AE: 0.5%
Winkler [17]	2018	744	Retrospective study cohort	primary THA/ revision THA	For the first time: cefazolin before the operation, 3 doses per repeat of preoperative VA after the operation, until the intraoperative culture is negative	Six months	6	Age: 66 ± 9, 65% Male, Diabetes (20%)	PJI: 4%, SSI: 1%, AE: 2.5%
Yang YZ [18]	2019	300	Retrospective study cohort	primary THA	2 g cefazolin before the operation	More than three months	4	Age: 64 ± 11, 55% Male, Hypertension (10%)	PJI: 2.8%, SSI: 0.3%, AE: 1%
Matziolis [19]	2020	8865	Retrospective study cohort	primary THA	2 g cefazolin before the operation	One year	4	Age: 67 ± 12, 55% Female, Obesity (18%)	PJI: 2%, SSI: 0.5%, AE: 1.2%
Iorio [20]	2020	4664	Retrospective study cohort	primary THA/ revision THA	Spectral antibacterial drugs were administered during the perioperative period	Three months	5	Age: 69 ± 14, 60% Male, Diabetes (22%)	PJI: 3%, SSI: 0.8%, AE: 1.5%

THA, total hip arthroplasty. The intervention measures were the prophylactic use of antibiotics in the wound during the operation in the experimental group, while no antibiotics were used in the wound during the operation in the control group

Risk of bias within studies

Using ROBINS-I indicated generally moderate risk due to confounding and selection, with limited reporting on attrition and outcome assessment; blinding was typically not applicable/unclear (Figs. 2 and 3).

Fig. 3.

Summary chart of risk bias

Results of meta-analysis

Periprosthetic infection (PJI)

All eight studies reported PJI. Heterogeneity was low (Chi² = 7.57, df = 7, P = 0.37; I² = 8%). The prespecified random-effects model showed lower PJI with intrawound + systemic vs. systemic alone (OR = 0.42, 95% CI: 0.30–0.58, P < 0.00001); fixed-effect results were consistent. As seen in Fig. 4.

Fig. 4.

Forest analysis chart of the overall incidence of PJI

Subgroup analysis: primary versus revision THA

Effects were consistent across surgery type with no detectable between-subgroup heterogeneity (I² = 0%). Pooled estimates showed reduced PJI in both primary THA and revision THA subgroups (OR = 0.30, 95% CI: 0.20–0.47, P < 0.00001; Fig. 5).

Fig. 5.

Forest plots of the incidence of PJI in patients undergoing primary and revised joint replacement surgeries

Primary THA results

Pooled estimate for primary THA: Intrawound antibiotics were associated with a significant reduction in PJI in primary THA (OR = 0.30, 95% CI: 0.20–0.47, P < 0.00001; I² = 0%).

Revision THA results

Revision THA was further divided into septic and aseptic revisions.

Septic revision THA

Pooled estimate for septic revision THA: In septic revision THA, intrawound antibiotics were associated with a significant reduction in PJI compared to systemic prophylaxis alone (OR = 0.28, 95% CI: 0.18–0.43, P < 0.00001; I² = 5%).

Aseptic revision THA

Pooled estimate for aseptic revision THA: In aseptic revision THA, intrawound antibiotics were also associated with a significant reduction in PJI (OR = 0.32, 95% CI: 0.21–0.47, P < 0.00001; I² = 8%).

Superficial surgical site infection (SSI)

Three studies reported superficial SSI. Heterogeneity was negligible (Chi² = 0.17, df = 2, P = 0.92; I² = 0%). The pooled effect was not significant (OR = 0.41, 95% CI: 0.10–1.66, P = 0.92; Fig. 6); analyses are underpowered.

Fig. 6.

Forest map of the total incidence of superficial infections

Adverse events (safety)

Reported in a limited subset of 3 studies. No significant difference was observed (OR = 0.89, 95% CI: 0.45–1.78, P = 0.75; Fig. 7); estimates are imprecise due to few studies.

Fig. 7.

Forest plot of the overall incidence of adverse reactions

Small-study effects and publication bias

Funnel plots were generated for PJI and secondary outcomes (Figs. 8, 9, 10 and 11). Visual inspection suggested approximate symmetry for most outcomes, with possible asymmetry for PJI. Given the limited number of studies (k = 8) and consequent low power, results of Egger’s test should be interpreted cautiously; findings did not rule out small-study effects. Exploratory trim-and-fill sensitivity analyses did not materially alter the direction of the pooled effect; full details are provided in the Supplement.

Fig. 8.

Funnel chart depicting the total PJI rate

Fig. 9.

Funnel plots depicting the frequency of periprosthetic joint infection in patients having initial and revision joint replacement operations

Fig. 10.

Funnel plot derived from the surface infection rate

Fig. 11.

Funnel diagram illustrating the frequency of adverse responses

Discussion

This meta-analysis of non-randomized, observational studies suggests that intrawound (local) antibiotic prophylaxis used as an adjunct to standard systemic perioperative antibiotics may be associated with lower PJI risk after THA, with consistent effects in primary and revision procedures and low heterogeneity for the primary outcome. In contrast, pooled estimates for superficial incisional SSI and adverse events were not significant; however, these analyses were based on few studies and are underpowered, yielding wide confidence intervals.

The effectiveness of intrawound antibiotics appears consistent across primary and revision THA procedures. However, the results of our analysis also suggest that the effectiveness of intrawound antibiotics may vary between septic and aseptic revision THA. In both septic and aseptic revisions, intrawound antibiotics showed a significant reduction in PJI compared to systemic antibiotics alone. This finding highlights the potential benefits of intrawound antibiotics not only in primary THA but also in complex revision procedures, where infection control is critical.

Several studies have explored the use of intrawound antibiotics in hip and knee arthroplasty, with mixed results. Some studies, including those by Kamran et al. (2022) and Abbaszadeh et al. (2025), have reported a significant reduction in periprosthetic joint infections (PJI) with the use of intrawound antibiotics, emphasizing their role in achieving high local antibiotic concentrations at the surgical site [21, 22]. However, other studies have raised concerns about the variability in protocols and the lack of high-quality evidence supporting their universal use [23]. In addition, alternative infection prevention strategies, such as Staphylococcus aureus decolonization, have shown promise in reducing infection rates. A systematic review by Bianco Prevot et al. (2024) concluded that Staphylococcus aureus decolonization protocols, including nasal decolonization, significantly reduced periprosthetic infection rates and should be considered as a standard practice in preoperative care [23]. Combining these strategies with intrawound antibiotics could further enhance infection control, especially in high-risk patients.

intrawound antibiotics may achieve high concentrations at the surgical site while limiting systemic exposure, offering a biologically plausible mechanism to reduce deep infection risk. Prior syntheses have reported mixed results some support a benefit of intrawound antibiotics in hip/knee arthroplasty [24, 25], whereas others (e.g., Wong et al. [26]) highlight insufficient evidence and gaps in safety data, so our findings should be interpreted cautiously given the observational evidence base.

Strengths include a comprehensive bilingual search (English and Chinese databases), PRISMA adherence, explicit primary vs. revision subgrouping, and a large pooled sample. We also applied tools appropriate for non-randomized evidence and examined small-study effects.

Several limitations temper certainty. First, all included studies were cohort designs, introducing risks of selection bias and residual confounding (e.g., patient risk profiles, perioperative care bundles, or surgeon preference). Second, intrawound protocols varied substantially (agent, dose, timing, and application technique), which constrains generalizability and precludes identification of an “optimal” regimen. Third, outcome definitions particularly for superficial SSI and follow-up durations were not fully standardized. Fourth, small-study effects and possible publication bias cannot be excluded; while sensitivity analyses did not materially change the direction of the effect, these methods have limited power with a small number of studies. Finally, only a small subset of studies reported superficial SSI and adverse events, limiting precision for safety inferences.

This localized effect may reduce the risk of infection without the systemic side effects that can result from prolonged systemic antibiotic use. Moreover, it may help mitigate antimicrobial resistance, a growing concern in perioperative care. The success of intrawound antibiotic prophylaxis may be attributed to the ability of intrawound antibiotics to achieve high concentrations at the surgical site while limiting systemic exposure, providing a biologically plausible mechanism to reduce the risk of deep infections such as PJI.

Although the number of studies included in this meta-analysis is limited, focusing on studies published since 2010 allowed us to include research that reflects more contemporary methodologies and standardized reporting. This ensures that our findings are relevant to current clinical practice and based on high-quality evidence. Future research with larger sample sizes and more randomized controlled trials is needed to further substantiate these findings.

The efficacy of intrawound antibiotics is likely due to their ability to achieve high local concentrations at the surgical site while limiting systemic exposure. This localized delivery of antibiotics allows for enhanced microbial killing in the area most susceptible to infection, without contributing to systemic side effects typically associated with prolonged antibiotic use. Additionally, intrawound antibiotics may help mitigate the growing concern of antimicrobial resistance, a critical issue in perioperative care. By focusing the antibiotics at the infection site, we may reduce the development of drug-resistant pathogens that can emerge from systemic antibiotic use. However, the variability in protocols (such as the choice of agent, dose, and timing) across studies complicates the interpretation of results and limits the ability to draw definitive conclusions about the most effective regimen.

Clinical implications should therefore be framed as associative rather than causal. Centers considering intrawound prophylaxis may (i) standardize local protocols alongside the routine systemic regimen, (ii) target higher-risk patients in whom potential benefit is greatest, and (iii) integrate antimicrobial stewardship and postoperative surveillance for adverse events and microbiological resistance. Importantly, intrawound prophylaxis complements, but does not replace rigorous aseptic technique and meticulous wound care. In terms of patient characteristics, the included studies presented a range of patient populations, with varying comorbidities and risk factors that could influence infection rates. Older patients, those with diabetes, and individuals with obesity are known to have an increased risk of infections after arthroplasty. The findings of this meta-analysis suggest that intrawound antibiotic prophylaxis may be an effective adjunct to standard systemic antibiotics in reducing PJI risk after THA. However, clinical applications should be framed cautiously, as the studies included in our analysis were non-randomized and observational. Centers considering the use of intrawound prophylaxis should standardize local protocols to align with routine systemic regimens, ensuring consistency in the application of antibiotics. Additionally, it is essential to target higher-risk patients, such as those with comorbidities like diabetes, obesity, or advanced age, who may benefit most from this additional infection control measure. Furthermore, integrating antimicrobial stewardship programs into perioperative care is critical to ensure that antibiotic use is both effective and responsible, minimizing the risks of resistance and adverse effects.

Future research priorities include multicenter, adequately powered randomized controlled trials that: (1) compare standardized intrawound regimens head-to-head (e.g., cefazolin adjuncts vs. intrawound vancomycin vs. no intrawound agent) on top of uniform systemic prophylaxis; (2) prespecify dose, timing, and application technique; (3) adopt harmonized diagnostic criteria for PJI/SSI with sufficient follow-up to capture late infections; and (4) include safety endpoints (e.g., nephrotoxicity, Clostridioides difficile infection), microbiology (pathogen profiles), antimicrobial resistance, and cost-effectiveness. Such trials are essential to move from association to causation and to define optimal, generalizable protocols. Future research should focus on well-designed, multicenter, randomized controlled trials to better assess the efficacy and safety of intrawound antibiotics in THA. Such studies should compare standardized intrawound regimens head-to-head, including different antibiotics such as cefazolin and vancomycin, in combination with uniform systemic prophylaxis. Additionally, it is crucial to prespecify the dose, timing, and method of application to reduce variability. Harmonized diagnostic criteria for PJI and SSI should be adopted across studies to ensure consistency in outcome reporting. Long-term safety, including monitoring for adverse events like nephrotoxicity and Clostridioides difficile infection, should be prioritized, along with microbiological surveillance for pathogen profiles and antimicrobial resistance. Cost-effectiveness analyses are also needed to evaluate the economic implications of intrawound antibiotic prophylaxis in different healthcare settings.

Conclusion

Limited evidence suggests that intrawound (local) antibiotic prophylaxis, when added to standard systemic perioperative antibiotics, may be associated with reduced PJI after both primary and revision THA. Evidence regarding superficial SSI and safety remains limited by few contributing studies and protocol heterogeneity. Rigorous, standardized randomized trials with longer follow-up are needed to define regimen-specific efficacy and long-term safety, including antimicrobial resistance.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: L.Z., X.X.; Data Curation: L.Z., Y.C., X.X.; Formal Analysis: L.Z., Y.C., X.X.; Funding Acquisition: X.X.; Investigation: L.Z., X.X.; Methodology: L.Z., X.X.; Project Administration: X.X.; Resources: L.Z., X.X.; Software: L.Z., Y.C., X.X.; Supervision: L.Z., Y.C., X.X.; Validation: L.Z., Y.C., X.X.; Visualization: L.Z., Y.C., X.X.; Writing—Original Draft: L.Z., Y.C.; Writing—Review & Editing: X.X.

Funding

This research received no external funding.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

All data were obtained from publicly available databases, and no additional ethics approval was required.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.