Non-pharmacological interventions for lactational mastitis: a systematic review and meta-analysis protocol Authors: Siyuan Zhang, Yuehan Ren, Qiong Zhang, Yingyi Fan Published: 2026 ABSTRACT AbstractBackgroundLactational mastitis is a common inflammatory condition that can adversely affect maternal well-being and is associated with early cessation of breastfeeding. Although antibiotics and analgesics are widely used in clinical practice, concerns regarding antimicrobial resistance and potential effects on the infant microbiome have prompted increasing interest in non-pharmacological approaches. A range of physical and biological interventions, including probiotics, therapeutic ultrasound, acupuncture and manual therapies, have been investigated; however, the comparative effectiveness and safety of these interventions remain unclear. This study aims to systematically evaluate the efficacy and safety of non-pharmacological interventions for lactational mastitis.Methods and analysisThis protocol is reported in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines. We will systematically search PubMed, Embase, Web of Science, the Cochrane Library, CNKI, Wanfang, VIP and SinoMed as well as relevant trial registries from database inception to March 2026.We will include randomised controlled trials enrolling adult lactating women diagnosed with mastitis. Eligible studies will compare non-pharmacological interventions with usual care, placebo or pharmacological treatments. Interventions involving pharmacologically active agents will be excluded.Two reviewers will independently perform study selection, data extraction and risk of bias assessment using the Cochrane RoB 2 tool. The primary outcome will be resolution of mastitis, defined as the return of body temperature to normal (<37°C) and resolution of localised breast symptoms (eg, pain, erythema and swelling). Secondary outcomes will include inflammatory markers (eg, C-reactive protein), pain intensity (measured using the Visual Analogue Scale or similar validated scales), incidence of adverse events, complete blood count parameters, time to symptom resolution or normalisation of body temperature, incidence of breast abscess and breastfeeding continuation rates.Meta-analysis will be conducted, where appropriate, using random-effects models. Continuous outcomes will be summarised using mean differences or standardised mean differences, and dichotomous outcomes using risk ratios, all with 95% CIs. Statistical heterogeneity will be assessed using the I² statistic. Subgroup and sensitivity analyses will be performed to explore potential sources of heterogeneity. Publication bias will be assessed using funnel plots and Egger’s test when sufficient studies are available. The certainty of evidence will be evaluated using the Grading of Recommendations Assessment, Development and Evaluation(GRADE)approach.Ethics and disseminationEthical approval is not required for this study as it is based on published data. The findings will be disseminated through peer-reviewed publications and conference presentations.PROSPERO registration numberCRD420261347640. STRENGTHS AND LIMITATIONS OF THIS STUDY A comprehensive search will be conducted across multiple English and Chinese databases (PubMed, Embase, CINAHL, CNKI, Wanfang and CBM) and trial registries to ensure broad literature coverage. A comprehensive search will be conducted across multiple English and Chinese databases (PubMed, Embase, CINAHL, CNKI, Wanfang and CBM) and trial registries to ensure broad literature coverage. By quantitatively synthesising the efficacy and safety of non-pharmacological interventions for lactational mastitis, this systematic review will address a current evidence gap and provide a reference for clinical practice and guideline development. By quantitatively synthesising the efficacy and safety of non-pharmacological interventions for lactational mastitis, this systematic review will address a current evidence gap and provide a reference for clinical practice and guideline development. Methodological rigour will be enhanced by following the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines, assessing risk of bias with the Cochrane RoB 2 tool and evaluating evidence certainty using the Grading of Recommendations Assessment, Development and Evaluation(GRADE)framework. Methodological rigour will be enhanced by following the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines, assessing risk of bias with the Cochrane RoB 2 tool and evaluating evidence certainty using the Grading of Recommendations Assessment, Development and Evaluation(GRADE)framework. Heterogeneity in intervention types (eg, manual therapies, probiotics and therapeutic ultrasound), outcome measures and treatment protocols may limit the feasibility of meta-analysis. Heterogeneity in intervention types (eg, manual therapies, probiotics and therapeutic ultrasound), outcome measures and treatment protocols may limit the feasibility of meta-analysis. Introduction Rationale Lactational mastitis is a prevalent condition that can range from moderate to severe in its clinical presentation, posing a substantial global health burden among breastfeeding women across diverse populations.1–4 The Academy of Breastfeeding Medicine has recently proposed a ‘mastitis spectrum’ framework, which conceptualises mastitis not merely as an acute infectious event but rather as a progressive condition involving localised milk stasis and subsequent tissue inflammation.5 6 However, definitions of mastitis vary. The WHO has defined mastitis as ‘an inflammatory condition of the breast, which may or may not be accompanied by infection’,7 and the diagnostic criteria remain debated.8 9 To address this ambiguity, this review will adopt a clear operational definition of lactational mastitis, requiring both localised breast symptoms (pain, erythema and swelling) and systemic features (fever ≥38.5°C or flu-like symptoms). The condition arises from a combination of milk stasis, nipple damage and altered breastmilk immune profiles.10 Multiple risk factors have been reported, including improper breastfeeding technique, breast trauma and maternal fatigue.11 During this inflammatory cascade, the profile of immune factors in breast milk is substantially altered, characterised by elevated levels of pro-inflammatory cytokines such as interleukin-8 and interferon-γ, alongside increased lactoferrin concentrations, contributing to localised pain, erythema, induration and systemic fever.12 13 Historically, the standard clinical management of lactational mastitis has relied heavily on narrow-spectrum antibiotics targeting Staphylococcus aureus and pharmacological analgesics.7 14 15 However, the evidence base for routine antibiotic use in mastitis remains limited.16 The indiscriminate prescription of antibiotics poses clinical concerns, including the risk of maternal antimicrobial resistance (AMR),17 potential disruptions to the infant’s developing microbiome18–20 and an associated risk of abrupt breastfeeding cessation.21 Consequently, there is increasing interest in alternative management strategies involving conservative, non-pharmacological interventions.22 Within Western healthcare settings, modern physical and biological therapies have gained significant traction. Several studies have shown that oral administration of specific probiotic strains (Lactobacillus spp.) may reduce the incidence of mastitis and alleviate related symptoms.23 24 Furthermore, topical applications of cold cabbage leaves, alongside physical cold compresses, are widely used to alleviate localised oedema and pain through pure physical cooling and mechanical fluid displacement.25–27 Therapeutic ultrasound has also been investigated for facilitating milk drainage, but randomised controlled trials have shown mixed results, with some reporting no benefit.28 29 Concurrently, Traditional Chinese Medicine and allied physical modalities offer a rich repository of non-pharmacological strategies. Acupuncture has been empirically validated in Western clinical trials for relieving inflammatory breast symptoms via mechanical neuromodulation.30 Moreover, targeted acupoint massage, traditional manual milk expression (tuina), and mechanical scraping (gua-sha) have demonstrated clinical efficacy in resolving ductal blockages and preventing abscess formation.31–33 Recent comprehensive analyses further confirm that non-invasive techniques such as acupressure, scraping and cupping can effectively improve lactation and relieve breast engorgement.34 Currently, a comprehensive systematic review quantitatively synthesising the safety of pure non-pharmacological interventions by the type of intervention (eg, acupuncture-related therapies, manual therapies, device-based therapies and probiotics) is lacking. Objectives The primary objective of this study is to systematically evaluate the efficacy and safety of global, non-pharmacological interventions for lactational mastitis. We aim to address the following PICO question: in lactating women with mastitis (P), do non-pharmacological interventions (I), compared with conventional care, placebo or pharmacological monotherapy (C), affect inflammatory markers, pain intensity and demonstrate a favourable safety profile (O)? Methods This protocol is reported in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 statement.35 36 The protocol has been prospectively registered in PROSPERO (registration number: CRD420261347640). Eligibility criteria Participants We will include studies involving adult lactating women (aged ≥18 years) diagnosed with lactational mastitis, defined as the presence of both localised breast symptoms (eg, pain, erythema, swelling and induration) and systemic features (eg, fever ≥38.5°C, flu-like symptoms such as myalgia or chills).37 We will exclude studies involving: Non-lactational mastitis. Non-lactational mastitis. Plasma cell mastitis or granulomatous mastitis. Plasma cell mastitis or granulomatous mastitis. Breast malignancy. Breast malignancy. Participants with severe systemic diseases. Participants with severe systemic diseases. Interventions We will include studies evaluating non-pharmacological interventions based on physical or biological mechanisms. We define non-pharmacological interventions as interventions that do not involve synthetic or conventional pharmacological agents, but may include biological therapies such as probiotics because they act through microbiome modulation rather than direct pharmacological mechanisms. These may include, but are not limited to: Manual therapies (eg, massage and milk expression). Manual therapies (eg, massage and milk expression). Device-based therapies (eg, therapeutic ultrasound). Device-based therapies (eg, therapeutic ultrasound). Microbiological therapies (eg, probiotics). Microbiological therapies (eg, probiotics). Acupuncture-related therapies. Acupuncture-related therapies. We specify ‘therapeutic ultrasound’ to distinguish it from diagnostic ultrasound, which will not be considered an intervention. In addition, interventions involving topical application of substances with known or potential pharmacological activity (eg, herbal poultices containing plant-derived compounds with anti-inflammatory or antimicrobial properties) will be excluded, as such interventions may exert effects through biochemical pathways rather than purely physical or mechanical mechanisms. Interventions involving pharmacologically active agents will be excluded, including: Herbal preparations containing plant-derived compounds with anti-inflammatory or antimicrobial properties and medicated patches. Herbal preparations containing plant-derived compounds with anti-inflammatory or antimicrobial properties and medicated patches. Antibiotic ointments. Antibiotic ointments. Non-steroidal anti-inflammatory drugs. Non-steroidal anti-inflammatory drugs. Comparators Comparators will include: Usual care or routine nursing care. Usual care or routine nursing care. Placebo or sham interventions. Placebo or sham interventions. Pharmacological treatments (eg, antibiotics or analgesics). Pharmacological treatments (eg, antibiotics or analgesics). Studies evaluating combined interventions will be included only if the incremental effect of the non-pharmacological component can be clearly assessed. Outcomes Studies will be included if they report at least one of the following outcomes. Primary outcomes: Resolution of mastitis, defined as the return of body temperature to normal (<37°C) and resolution of localised breast symptoms (eg, pain, erythema and swelling). Resolution of mastitis, defined as the return of body temperature to normal (<37°C) and resolution of localised breast symptoms (eg, pain, erythema and swelling). Secondary outcomes: Inflammatory markers (eg, C-reactive protein).38 Inflammatory markers (eg, C-reactive protein).38 Pain intensity (measured using the Visual Analogue Scale or similar validated scales). Pain intensity (measured using the Visual Analogue Scale or similar validated scales). Incidence of adverse events. Incidence of adverse events. Complete blood count parameters. Complete blood count parameters. Time to symptom resolution or normalisation of body temperature. Time to symptom resolution or normalisation of body temperature. Incidence of breast abscess. Incidence of breast abscess. Breastfeeding continuation rates. Breastfeeding continuation rates. Study design We will include only randomised controlled trials. We will exclude: Non-randomised studies. Non-randomised studies. Observational studies. Observational studies. Case reports or case series. Case reports or case series. Reviews, editorials and conference abstracts without sufficient data. Reviews, editorials and conference abstracts without sufficient data. Information sources and search strategy A comprehensive search will be conducted in PubMed, Embase, Web of Science, the Cochrane Central Register of Controlled Trials (CENTRAL) and four Chinese databases (CNKI, Wanfang Data, VIP Database and SinoMed) from inception to March 2026. The search strategy will combine controlled vocabulary terms and free-text terms related to lactational mastitis and non-pharmacological interventions. The search strategy will be adapted for each database. The full search strategies are provided in the online supplemental material. Study selection and data extraction All records will be imported into reference management software and duplicates will be removed. Two reviewers (YR and SZ) will independently screen titles and abstracts, followed by full-text assessment of potentially eligible studies. Disagreements will be resolved through discussion or consultation with a third reviewer (YF). The study selection process will be presented in a PRISMA flow diagram (figure 1). Flowchart of the literature review and selection process. This figure presents the pre-specified stages of record identification, screening, eligibility assessment and inclusion. Numbers will be updated on completion of the review. Data extraction will be conducted independently by two reviewers (YR and SZ) using a pre-piloted standardised form. Extracted data will include study characteristics, participant details, intervention characteristics and outcomes, including adverse events. For missing or unclear data, we will attempt to contact the corresponding authors. Where necessary, missing SD will be calculated from available statistics (eg, SEs or CIs) in accordance with the Cochrane Handbook. Risk of bias assessment Risk of bias will be assessed independently by two reviewers using the Cochrane RoB 2 tool.39 Data synthesis and statistical analysis Where appropriate, meta-analysis will be performed using random-effects models, considering the anticipated clinical heterogeneity across interventions. Continuous outcomes will be summarised using mean differences or standardised mean differences, and dichotomous outcomes using risk ratios, all with 95% CIs. Statistical heterogeneity will be assessed using the I² statistic, with values of 25%, 50% and 75% representing low, moderate and high heterogeneity, respectively.40 Where substantial heterogeneity is identified (I² >50%), potential sources will be explored through subgroup analyses and sensitivity analyses. Predefined subgroup analyses will include: Type of intervention (eg, acupuncture-related therapies, manual therapies, device-based therapies and probiotics). Type of intervention (eg, acupuncture-related therapies, manual therapies, device-based therapies and probiotics). Timing of intervention (eg, early vs delayed treatment). Timing of intervention (eg, early vs delayed treatment). Sensitivity analyses will be conducted by excluding studies at high risk of bias and by applying alternative statistical models. If at least 10 studies are included in a meta-analysis, publication bias will be assessed using funnel plots and Egger’s test. If quantitative synthesis is not appropriate due to substantial clinical or methodological heterogeneity, a narrative synthesis will be conducted. Findings will be summarised descriptively according to intervention type, outcome measures and study characteristics. Confidence in cumulative evidence The certainty of evidence will be assessed using the GRADE approach.41 42 Patient and public involvement statement Patients or members of the public were not involved in the design, conduct, reporting or dissemination of this study. Data availability statement No data are available as this is a protocol for a systematic review. Ethics and dissemination Ethical approval is not required for this study as it is based on published data. The findings will be disseminated through peer-reviewed publications and conference presentations. Study timeline Protocol registration: March 2026 (completed). Protocol registration: March 2026 (completed). Literature search: May–June 2026. Literature search: May–June 2026. Study selection: June–July 2026. Study selection: June–July 2026. Data extraction: July–August 2026. Data extraction: July–August 2026. Risk of bias assessment: August 2026. Risk of bias assessment: August 2026. Data analysis: September–October 2026. Data analysis: September–October 2026. Manuscript preparation: November–December 2026. Manuscript preparation: November–December 2026. Planned submission: December 2026. Planned submission: December 2026. Discussion This systematic review and meta-analysis will provide a comprehensive and methodologically rigorous synthesis of the available evidence regarding the efficacy and safety of non-pharmacological interventions for lactational mastitis. By focusing specifically on interventions that operate through physical or biological mechanisms, this study aims to generate clinically relevant evidence to inform antibiotic-sparing strategies in the management of this common postpartum condition. Given the increasing concerns surrounding AMR and the potential disruption of the infant microbiome associated with antibiotic use, there is a growing need to identify effective alternative therapies. By systematically comparing a broad range of non-pharmacological interventions, this study is expected to clarify their relative effectiveness and safety profiles, thereby supporting more individualised and evidence-based clinical decision-making. This study has several methodological strengths. First, the protocol is developed in accordance with PRISMA-P guidelines and incorporates a comprehensive search strategy across both English and Chinese databases, which is expected to maximise evidence coverage and minimise publication bias. Second, rigorous methodological approaches, including duplicate study selection, standardised data extraction, risk of bias assessment using the Cochrane RoB 2 tool and evaluation of evidence certainty using the GRADE framework, will enhance the transparency, reliability and interpretability of the findings. Nevertheless, several limitations should be acknowledged. Substantial clinical and methodological heterogeneity is anticipated due to variations in intervention types, treatment intensity, practitioner expertise and outcome measurement across studies. Such heterogeneity may complicate data synthesis and interpretation, even with the use of random-effects models. Second, blinding is often challenging in trials of physical or manual interventions, which may introduce performance and detection bias, particularly for subjective outcomes such as pain. The overall quality of included trials may also vary, potentially influencing the certainty of evidence. To address these issues, we plan to conduct subgroup and sensitivity analyses to explore sources of heterogeneity and assess the robustness of the findings. Supplementary material