Restoring skeletal remodeling in antiresorptive-treated patients: clinical outcomes of teriparatide in medication-related osteonecrosis of the jaw

Abstract

Medication-related osteonecrosis of the jaw (MRONJ) is a challenging complication associated with antiresorptive and antiangiogenic therapies. Prolonged suppression of bone remodeling has been implicated as a key mechanism in its pathophysiology. Teriparatide, an osteoanabolic agent that restores coupled bone remodeling, has been proposed as a potential therapeutic strategy. This review aimed to synthesize reported clinical outcomes of teriparatide use in MRONJ within a systemic remodeling-based framework. We conducted a structured descriptive review based on a systematic literature search and reported in accordance with PRISMA 2020. Studies reporting therapeutic use of teriparatide in established MRONJ were included regardless of the triggering medication. Data were extracted on patient characteristics, MRONJ stage, lesion site, antiresorptive exposure, teriparatide regimen, antiresorptive interruption, antiresorptive type, and clinical outcomes. Because of substantial heterogeneity in study design, patient populations, co-interventions, and outcome reporting, evidence was synthesized qualitatively. Thirty-four publications including 207 patients were included. Clinical outcomes were available for 201 patients: complete coverage/healing was reported in 163, partial improvement in 24, no change in 12, and worsening in 2. Teriparatide discontinuation or dropout after treatment initiation was reported in 13 patients. Stage 2 disease predominated among stageable cases, mandibular involvement was the most frequent anatomical site, and daily teriparatide dosing was more commonly reported than weekly regimens. Bisphosphonate exposure predominated, whereas denosumab-associated cases were fewer; therefore, differential effectiveness by antiresorptive class could not be determined. Published reports describe frequent clinical improvement after teriparatide use in MRONJ. However, the evidence remains descriptive, heterogeneous, and commonly confounded by concomitant therapies. These findings support a biologically plausible remodeling-based interpretation but do not establish efficacy or define teriparatide as standard therapy.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00198-026-08150-y.

Introduction

Medication-related osteonecrosis of the jaw (MRONJ) is a severe adverse effect most commonly associated with antiresorptive and antiangiogenic therapies [1–4]. Although antiresorptive agents represent the most frequent triggers, MRONJ has also been associated with other systemic therapies. It significantly impairs patients’ quality of life and remains difficult to manage, with variable therapeutic responses and frequent local persistence [5–7]. Current treatment strategies often involve temporary or permanent suspension of antiresorptive agents; however, this approach may increase fracture risk in osteoporotic patients and, in oncologic settings, potentially compromise skeletal protection and disease control [8, 9]. Therefore, additional strategies that promote jawbone healing without undermining systemic skeletal stability are needed.

Although the pathophysiology of MRONJ is multifactorial, prolonged suppression of bone remodeling is considered a central contributing mechanism. The jaws represent a unique skeletal environment because of their high turnover rate, continuous exposure to the oral microbiome, and frequent microtrauma related to mastication and dental procedures [10–14]. In this context, MRONJ may be interpreted not only as a localized maxillofacial complication but also as a site-specific manifestation of impaired skeletal remodeling in an anatomically high-demand region.

From this perspective, therapeutic strategies aimed at restoring physiological remodeling may offer a complementary approach to conventional local management. Teriparatide, a recombinant human parathyroid hormone analog, stimulates bone formation and can reactivate coupled remodeling dynamics [15–19]. This provides a biologically plausible rationale for its potential role in MRONJ management, particularly in patients with osteoporosis and prolonged antiresorptive exposure. However, in the available reports, teriparatide was generally administered as a targeted intervention for osteonecrosis rather than as part of a comprehensive systemic skeletal management strategy.

The use of teriparatide also requires careful interpretation in oncologic contexts. Antiresorptive therapy in cancer patients may be prescribed either for prevention of skeletal-related events in metastatic disease or multiple myeloma, or for cancer treatment-induced osteoporosis in the absence of metastatic bone disease. These scenarios differ in cumulative exposure, treatment intent, and clinical risk. Because the included reports did not consistently specify metastatic status, dosing regimen, or oncologic disease activity, conclusions regarding malignancy-associated MRONJ must remain cautious.

Several clinical reports have described clinical improvement following teriparatide use in patients with MRONJ, particularly in moderate or advanced stages. However, the available evidence remains heterogeneous in design and quality, and its overall strength has not been clearly contextualized within a systemic remodeling-based interpretation of the disease.

The primary objective of this review was to descriptively synthesize reported clinical outcomes of teriparatide use in MRONJ. A secondary objective was to discuss whether the observed clinical patterns are biologically consistent with a remodeling-based interpretation of MRONJ pathophysiology and management (Fig. 1).

Fig. 1.

Teriparatide and remodeling restoration in MRONJ

Materials and methods

Study design

This study was designed as a structured descriptive review based on a systematic literature search and qualitative evidence synthesis. PRISMA 2020 [20] was used as a reporting framework for study identification, selection, and transparent presentation of the review process, but the objective was not to perform a comparative effectiveness systematic review or meta-analysis (Fig. 2).

Fig. 2.

PRISMA 2020 flow diagram showing the identification, screening, eligibility assessment, and inclusion of studies evaluating teriparatide for medication-related osteonecrosis of the jaw (MRONJ). Duplicate records were identified and removed using Mendeley before title and abstract screening

Search strategy

We performed a systematic search covering the period from January 2007 to March 2025 using the following electronic databases: PubMed/MEDLINE, SciELO, Cochrane Library, Trip Database, and Google Scholar. Gray literature sources were also screened to reduce publication bias.

The search strategy combined Medical Subject Headings (MeSH) terms and Boolean operators related to teriparatide and MRONJ. Core search expressions included combinations of:

• “Teriparatide,”

• “Parathyroid Hormone,”

• “Medication-Related Osteonecrosis of the Jaw,”

• “Osteonecrosis of the Jaw,”

• “Osteoporosis,”

• “Denosumab,”

• “Romosozumab,”

• “Multiple Myeloma,”

• “Osteomyelitis.”

Representative Boolean combinations included:

• “Teriparatide AND Osteonecrosis,”

• “Teriparatide AND MRONJ,”

• “Teriparatide AND Denosumab AND Osteonecrosis,”

• “Teriparatide AND Romosozumab AND Osteoporosis,” and related permutations.

The database search identified 42,657 records. Duplicate references were removed using Mendeley reference management software.

Study selection

After duplicate removal, titles and abstracts were screened for relevance to the therapeutic use of teriparatide in patients diagnosed with MRONJ. Following this preliminary screening, 1853 records were retained for further evaluation.

A total of 147 full-text articles were independently reviewed by two authors. Disagreements regarding study eligibility were resolved through discussion and, when necessary, consultation with a third reviewer.

Studies were eligible if they:

• Reported original clinical cases of patients diagnosed with MRONJ or bisphosphonate-related osteonecrosis of the jaw,

• Included teriparatide or recombinant human parathyroid hormone (rhPTH (1–34)) as part of treatment, either as monotherapy or adjunctive therapy,

• Provided extractable clinical outcome data related to bone exposure evolution,

• Included sufficient methodological description to allow data extraction.

Eligible study designs included randomized clinical trials, nonrandomized trials, cohort studies, case-control studies, case series, case reports, and letters or correspondence when they contained original extractable patient-level clinical data.

Studies were excluded if they were:

• Literature reviews, systematic reviews, or meta-analyses,

• Animal or preclinical studies,

• Clinical guidelines or management protocols without primary patient data,

• Studies involving patients younger than 20 years of age (due to ongoing skeletal modeling).

Reports were also excluded when treatment regimen or outcome data could not be reliably extracted or harmonized with the review framework.

Harper and Fung published the earliest report describing the use of recombinant human parathyroid hormone (rhPTH(1–34)) in bisphosphonate-associated osteonecrosis of the mandible in 2007. This publication is recognized as the first description of this therapeutic approach. However, it was not included in the quantitative analyses or outcome tables because the reported dosing regimen was subsequently corrected, precluding reliable harmonization with the teriparatide treatment protocols extracted from the remaining studies. Therefore, the report is discussed as historical background but was excluded from the quantitative synthesis to maintain methodological consistency.

After applying inclusion and exclusion criteria, 34 publications were included in the final qualitative synthesis [21–54]. Given the predominance of case reports, small case series, and heterogeneous interventional contexts, the review was designed to summarize reported clinical patterns rather than estimate a pooled treatment effect.

Data extraction

Data were independently extracted from each included study using a predefined standardized template.

The following variables were collected when reported:

MRONJ-related variables:

• Number of patients

• Age and sex

• Underlying indication for antiresorptive therapy (osteoporosis or oncologic condition)

• Route of administration of the antiresorptive or triggering agent

• Duration of antiresorptive therapy prior to MRONJ diagnosis

• Time from antiresorptive initiation to MRONJ diagnosis

• MRONJ stage according to the AAOMS (Ruggiero) classification

• Anatomical site (mandible, maxilla, or both)

• Procedure associated with MRONJ onset

• Presence of superimposed infection or chronic osteomyelitis and corresponding management

Systemic and behavioral risk factors when reported, including glucocorticoid or corticosteroid use, cigarette smoking, alcohol consumption, diabetes mellitus, rheumatoid disease or rheumatoid arthritis, cardiovascular disease, and other relevant comorbidities or concomitant medications.

Teriparatide-related variables:

• Time from MRONJ diagnosis to teriparatide initiation

• Teriparatide regimen (daily or weekly)

• Dosage and frequency

• Duration of teriparatide therapy

• Antiresorptive interruption status at teriparatide initiation

• Concomitant conservative or surgical management

• Imaging changes associated with therapy

• Changes in bone turnover markers

• Reported adverse events and teriparatide discontinuation or dropout when available

• Assessment of quality of life when available

When variables were inconsistently reported across studies, data were recorded as “not reported” (N/R).

Although a broad set of clinical and treatment-related variables was extracted to comprehensively characterize the reported cases, the primary objective of this review was to synthesize clinical outcomes associated with teriparatide use in MRONJ. Therefore, descriptive subgroup analyses were limited to variables that were consistently reported across a sufficient number of cases to allow meaningful interpretation. Variables with highly sparse or inconsistent reporting were documented descriptively but were not included in quantitative summaries.

Variables were extracted to comprehensively characterize the clinical context of published cases; however, extraction of a variable did not imply that it was part of the primary analytic framework of this review. The primary analytic emphasis remained on reported clinical evolution after teriparatide use.

Outcome definition

The primary outcome was the clinical evolution of exposed bone following teriparatide therapy, categorized as follows:

• Complete healing: full mucosal coverage of previously exposed bone

• Clinical improvement: partial reduction in the size of exposed bone

• No change

• Worsening

The denominator for outcome analysis corresponded to all patients for whom clinical evolution was explicitly reported, regardless of whether teriparatide therapy was completed.

Outcome categories were harmonized across reports based on the authors’ descriptions of clinical evolution, as no uniform outcome instrument was used across the included studies.

Quality assessment and risk of bias

The NICE framework [55] was used to contextualize the level of evidence of included therapeutic studies. For randomized clinical trials, risk of bias was additionally assessed using the Cochrane tool. Because most included studies were case reports, case series, letters/correspondence with original clinical data, or small observational studies, a formal domain-based risk-of-bias comparison across all study types was not feasible; instead, major sources of bias were considered narratively during interpretation.

Data synthesis

Given the heterogeneity in study design, reporting standards, patient populations, teriparatide regimens, and concomitant interventions, quantitative meta-analysis was not performed.

Descriptive subgroup analyses were conducted for variables reported in a sufficient number of cases, including:

• MRONJ stage

• Anatomical site

• Duration of antiresorptive exposure

• Antiresorptive interruption status

• Teriparatide dosing regimen

• Type of antiresorptive exposure

For each variable, proportions were calculated using the number of patients for whom the specific variable was explicitly reported (“reported subset”), to avoid distortion due to missing data. Antiresorptive type was summarized descriptively because several patients had mixed or sequential exposure, and outcomes were not consistently stratified by drug class.

Results

Study selection and cohort overview

A total of 34 publications comprising 207 patients with MRONJ treated with teriparatide were included. Descriptive baseline characteristics were summarized using all reported patients. Clinical outcomes were analyzed only in the subset of 201 patients for whom post-treatment clinical status was available. Demographic and MRONJ characteristics are reported in Table 1, and teriparatide treatment and clinical outcomes are reported in Table 2. Six patients were not evaluable for the final clinical outcome or had incomplete post-treatment outcome data; therefore, denominators differ across variables according to data availability in the original reports.

Table 1.

Demographic and MRONJ characteristics. The table summarizes baseline characteristics from the 207 patients included in the extracted reports. Percentages are based on variable-specific reported subsets, because several variables were incompletely or heterogeneously reported across studies

Author/year of publication	Type of study	Patients (n)	Age in years	Gender	Reason for medication use	Comorbidities/medications	Antiresorptive	Duration of antiresorptive exposure before MRONJ development	MRONJ Area	MRONJ stage according to AAOMS (patients)	Trigger
Yoshiga D et al. 2021	Prospective cohort study	Daily (9) - Weekly (9)	Daily 77–86 Mean: 81.8 (9) - Weekly 64–88 Mean: 76.8 (9)	Daily: F (8); M (1) - Weekly: F (7); M (2)	OP (18)	Daily: DM (2); CT (1); N/R (3) - Weekly: DM(2)	Daily: AL (4); RS (2); MN (1); N/R (1); more than 2 BP (1) - Weekly: AL (8); RS (1)	Daily: 3.4–17Y Mean: 6.2Y (9) - Weekly: 1.0–8.0Y Mean: 4.4Y (9)	Daily: MX (1); MD (5); N/R (3) - Weekly: MX (1); MD (8)	Daily: 2 (4); 3 (2); N/R (3) - Weekly: 2 (7); 3 (2)	N/R (18)
Kim KM et al. 2023	Case-control	Daily (17) - Weekly (12)	Daily Mean: 76.41 (17) - Weekly Mean: 75.50 (12)	Daily: F (16); M (1) - Weekly: F (12)	OP (29)	Daily: DM (5); CT (4); ALH (1); B (1); CD (4) - Weekly: DM (4); CT (3); ALH (0); TB (0); CD (3)	Daily: AL (6); IB (2); RS (1); ZL (4); DB (4) Weekly: AL (4); IB (2); RS (1); ZL (5); DB (0)	Daily Mean: 59.65; M (17) - Weekly Mean: 57 M (12)	Daily: MX (9); MD (8) - Weekly: MX (5); MD (7)	Daily: 1 (4); 2 (10); 3 (3) - Weekly: 1 (2); 2 (5); 3 (5)	Daily: EXO (8); PD (2); DN (2); IP (1); SP (4) - Weekly: EXO (5); PD (1); DN (1); IP (2); SP (3)
Mariño RQ et al. 2015	Case report	1	80	F	OP	N/R	AL	8Y	MX	3	EXO
Mohamed MMS et al. 2022	Case report	1	81	F	OP	CD	AL	3Y	MD	3	EXO
Sim IW et al. 2020	Placebo-controlled randomized double-blind clinical trial	15	59–71 Mean: 64 (15)	F (7); M (8)	MM (10); breast CA (1); prostate CA (2); OP (2)	DM (2); ALH (8); TB (13); CT (9)	N/R	< 1Y (9) > 1Y (6)	1 site N/R (10); ≥ 2 sites N/R (5)	0–1 (6); 2 (7); 3 (2)	EXO (7); N/R (8)
Morishita K et al. 2020	Retrospective multicenter case series	29	56–88 Mean: 77.4 (29)	F (23); M (6)	N/R (29)	CT (8); DM (2)	AL (13); RS (3); MN (2); DB (2); ZL (1); N/R (8)	1-4Y (29)	MX (4); MD (22); MX + MD (3)	1 (1); 2 (21); 3 (7)	EXO (16); PD (5); N/R (8)
Javelot MJ et al. 2020	Case report	1	69	F	OP	RA–hydroxychloroquine–MT–infliximab–leflunomide–CT	RS 5 mg/day—14 months + RTX 1 g two applications (2 years after RS)	1-4Y	MX + MD	2	SP
Ohbayashi Y et al. 2020	Randomized control clinical trial	Daily (6) - Weekly (7)	Daily: 78–85 Mean: 82 (6) - Weekly: 79–82 Mean: 82.5 (7)	F (13)	OP (13)	Daily: RA(2); CT(3); MT (2); infliximab (1) - Weekly: RA(3); CT(2); cyclosporine (1)	Daily: RS (3), AL(3) - Weekly: RS (3), AL (1), MN (2), N/R(1)	Daily: > 4Y (6) - Weekly: > 4Y (7)	Daily: MD (5); MX(1) - Weekly: MX(2), MD (3); MX + MD(1); N/R (1)	Daily: 2 (5); 3 (1) - Weekly: 2 (3); 3 (3); N/R(1)	N/R (13)
Kim JY et al. 2019	Case report	1	85	F	OP	N/R	RS	4Y	MD	3	IP
Pispati A et al. 2018	Case report	1	60	F	OP	AHT–DM	AL	4Y	MD	N/R	N/R
Mizohata K et al. 2018	Case report	1	78	F	OP	DM, AHT, colon CA, pancreatic CA, lumbar FX	RS	7.5Y	MX	2	SP
Jung J et al. 2017	Non-randomized clinical trial	6	59–86	F (6)	OP (6)	N/R (6)	N/R (6)	N/R (6)	N/R (6)	N/R (6)	N/R (6)
Zushi Y et al. 2017	Case report	1	66	F	OP	RF	AL	6Y	MD	N/R	IP
Yao M et al. 2016	Case report	1	87	F	OP	N/R	MN	2Y	MD	3	EXO
Kakehashi H et al. 2015	Case series	10	63–91 Mean: 79 (10)	F(10)	OP (10)	RA (1); DM (3); CT (5)	AL (4); RS (2); more than 2 BP (3); MN (1)	1- 4Y (5) > 4Y (5)	MX + MD (1); MX (1); MD (8)	2 (7); 3 (3)	EXO (6); PD (4)
Yamachika E et al. 2015	Case report	1	81	M	OP	N/R	MN	2Y	MD	3	EXO
Doh RM et al. 2015	Case report	1	75	F	OP	AHT, HCL	AL	2.5Y	MD (1)	N/R	EXO
Pelaz A et al. 2014	Non-randomized clinical trial	4	64–84 Mean: 73.5 (4)	F(4)	OP (4)	CT (2); AHT (1); anxiolytics (1); RA (2)	AL (2); IB (2)	> 4Y (4)	MD (4)	3 (4)	EXO (2); SP (2)
García B et al. 2014	Case report	1	84	F	OP	AHT	DB 60 mg	5 M	MD	N/R	IP
Neuprez A et al. 2014	Case report	1	58	M	OP	N/R	DB	4 M	MD	N/R	N/R
Kim KM et al. 2014	Prospective cohort study	15	Mean: 77.1 (15)	F (14); M (1)	OP (15)	N/R (15)	AL (6); RS (6); IB (1); ZL (1); PM (1)	2–6Y Mean: 3Y (15)	MD (11); MX (4)	2 (14); 3 (1)	IP (5) EXO (7) SP (3)
Yoshiga D. et al. 2013	Case report	2	87, 81	F:2	N/R (2)	N/R (2)	AL Wk 70 μg (1), 35 μg (1)	4 Y; 5 Y	MD (1); MX (1)	3 (2)	EXO (1) SP (1)
Ohbayashi Y et al. 2013	Case report	1	78	F	OP	DM, AHT, dysuria, OP	AL	> 4Y	MD	3	EXO
Kwon YD et al. 2012	Case series	6	68–88 Mean: 77.5 (6)	F(4) M(2)	N/R (6)	DM (3); RA (1); AHT (1); N/R (2)	AL (4); RS (1); more than 2 BP (1)	1-4Y (2); > 4Y (4)	MD (5); MX (1)	2 (4); 3 (2)	N/R (6)
Iwamoto J et al. 2012	Case report	1	79	F	OP	CVA, GU	RS 2.5 mg/D (6Y), 17.5 mg/Wk (7 M)	6Y	MD	N/R	EXO
Choi et al. 2024	Case series	3	68–87 Mean: 76 (3)	F(2) M(1)	OP (3)	N/R (2), DM (1)	AL (1); IB (1); more than 2 BP (1)	1–4 Y (2); > 4 Y (1)	MD (2); MX (1)	3 (2); 2 (1)	N/R (1); PD (1); EXO (1)
Kim et al. 2024	Case series	2	78–81 Mean: 79.5 (2)	F(2)	OP (2)	DM (1), AHT (1), N/R (1)	N/R (1); more than 2 BP (1)	4 Y(1); 5 Y (1)	MD (1); MX (1)	3 (2)	EXO (1); IP (1)
Singh et al. 2025	Case report	1	72	F	CA	AHT, OP, hyperlipidemia, atrial fibrillation	More than 2 BP	> 4 Y	MD	2	EXO
Park JH et al. 2025	Retrospective cohort study	Conservative (8) - Surgical (27)	Conservative Mean 74.75 (8) - Surgical Mean 77.11 (27)	Conservative F (8) - Surgical F(27)	OP (35)	Conservative: DM (3); CD (2); AHT (6); RA (1); Others (3) - Surgical DM (6); CD (2); AHT (15); RA (1); Others (7)	Conservative: AL (3); IB (2); RS (1); N/R (2) - Surgical AL (9); IB (9); RS (2); PM (1); N/R (6)	Conservative: More than 4 Y (8) - Surgical: More than 4 Y (27)	Conservative: MX (5); MD (3) - Surgical: MX (6); MD (21)	Conservative 1 (4); 2 (3); 3 (1) - Surgical 1 (5), 2 (19), 3 (3)	Conservative EXO (3); IP (3); SP (2) - Surgical EXO (16); IP (7); SP (4)
Cheung & Seeman, 2010	Case report (letter to the editor)	1	88	F	OP	Asthma	AL	10 Y	MD	N/R	EXO
Narongroeknawin et al. 2010	Case report	1	63	M	OP	Erosive osteoarthritis	AL	5 Y	MX	N/R	EXO
Lee et al. 2011	Case report	1	78	F	OP	FX	AL	3 Y	MD	3	EXO
Lau AN & Adachi JD. 2009	Case report/letter to the editor	1	56	F	OP	OA	ZL	1–4 Y	MX + MD	2	EXO
Tsai KY et al. 2010	Case report/letter to the editor	1	72	F	OP	N/R	AL	1–4 Y	MD	3	IP
Total	Case report (18); case report/letter to the editor (3); prospective cohort study (2); non-randomized clinical trial (2); randomized clinical trial (2); case series (5); case-control study (1); retrospective cohort study (1)	207	56–91	F (183); M (24)	OP (156); MM (10); Prostate CA (3); Breast CA (1); N/R (37)	CT (38); DM (37); RD/RA (12); TB (14); AHT (30); ALH (9); OA: (1), N/R (41); CD (12); Others (23)	AL (80); RS (30); MN (8); DB (8); ZL (12); IB (19); PM (2); more than 2 BP (8); N/R (40)	Less than 1 Y (11); Between 1 to 4 Y (100); more than 4 Y (90); N/R (6)	MX + MD (8); MX (46); MD (128); multiple sites (≥ 2) (5); N/R (20)	Stages 0–1 (6); Stage 1 (16); Stage 2 (114); Stage 3 (53); N/R(18)	EXO (83); IP (23); PD (13); SP (21); DN (3); N/R (64)

F female, M male, OP osteoporosis, MM multiple myeloma, CA cancer, DM diabetes mellitus, CT corticosteroid, TB tobacco consumption, ALH alcohol consumption, CD cardiovascular disease, RA rheumatoid arthritis, OA osteoarthritis, MT methotrexate, AHT arterial hypertension, HCL hypercholesterolemia, FX fracture, RF renal failure, CVA cerebrovascular accident, GU gastric ulcer, AL alendronate, RS risedronate, MN minodronate, IB ibandronate, ZL zoledronate, DB denosumab, RTX rituximab, PM pamidronate, Wk weekly, EXO exodontia, PD periodontal disease, DN denture, SP spontaneous, IP implants-peri implantitis, AHG acute herpetic gingivostomatitis, MX maxilla, MD mandible, MX + MD maxillomandibular, Y years, W weeks, D days

Table 2.

Teriparatide treatment and clinical outcomes. Outcome analysis was based on 201 patients with reported post-treatment clinical status among 207 included patients. Patients who discontinued or dropped out after receiving teriparatide were retained in the evaluable outcome dataset when post-treatment clinical outcome data were available. This table also includes the NICE evidence classification assigned to each included study

Author/year of publication	Patients (N = 207; evaluable n = 201)	Antiresorptive therapy halted before/during PTH treatment	Time from MRONJ diagnosis to PTH initiation	PTH treatment duration (months)	PTH Dosage	Treated only with PTH	Therapy prior to PTH treatment
							Conservative/antibiotic
Yoshiga D et al. 2021	Daily (6) - Weekly (9)	N/R (15)	N/R (15)	Daily: 3.7 (6) - Weekly: 3.0 (9)	Daily 20 µg (6) - Weekly 56.5 µg (9)	No	AB (Penicillin) + CLH + AI
Kim KM et al. 2023	Daily (17) - Weekly (12)	Yes (29)	N/R	Daily: 3.6 (17) - Weekly: 3 (12)	Daily 20 µg (17) - Weekly 56.5 µg (12)	No	AB + CLH + LIR
Mariño RQ et al. 2015	1	-	-	2	20 µg	No	AB (amoxicillin + Ac Clavulanate) + CLH
Mohamed MMS et al. 2022	1	-	-	24	20 µg	No	14 months AB + CLH + DBD
Sim IW et al. 2020	15	No (15)	N/R (15)	2 (15)	Daily 20 µg (15)	No	-
Morishita K et al. 2020	22 (29*)	Yes (29)*	N/R (29)*	> 12 M (22); N/R (7)*	Daily 20 ug (29)	Yes (2)	-
Javelot MJ et al. 2020	1	Yes	4 M	9	Daily 20 µg	No	AB (amoxicillin 2 g/12H for 2W) + CLH + PT (every 2 weeks)
Ohbayashi Y et al. 2020	Daily (6) - Weekly (6)	No (12)	N/R (12)	6 (12)	Daily 20 µg (6) - Weekly 56.5 µg (6)	No (12)	-
Kim JY et al. 2019	1	Yes	2 M	8	Weekly 56.5 µg	No	AB (N/R) + CLH + AI
Pispati A et al. 2018	1	Yes	N/R	18	Daily 20 µg	No	-
Mizohata K et al. 2018	1	Yes	3 M	18	Weekly 56.5 µg	Yes	-
Jung J et al. 2017	6	Yes (6)	N/R (6)	1–4 (6)	Daily 20 µg (6)	No (6)	-
Zushi Y et al. 2017	1	Yes	3 M	5	Daily 20 µg	No	AB OA (meropenem–ampicillin + sulbactam–clarithromycin) + MW
Yao M et al. 2016	1	Yes	1 M before	18	Weekly 56.5 µg	No	LIR
Kakehashi H et al. 2015	8	Yes (8)	N/R (8)	1–6 (1); 7–12 (4); > 1 Y (3)	Daily 20 µg (8)	No (8)	-
Yamachika E et al. 2015	1	N/R	5 W	3	Weekly 56.5 µg	No	AB (amoxicillin 750 mg) + LIR
Doh RM et al. 2015	1	Yes	At the time of diagnosis, treatment is suspended and TPTD is initiated	6	Daily 20 µg	No	AB (penicillin N/R) + CLH + LIR
Pelaz A et al. 2014	4	N/R (4)	N/R (4)	1–6 (2); 7–12 (2)	Daily 20 µg (4)	No	Treatment proposed by the AAOMS for MRONJ (patient 1 and 2 SCT)
García B et al. 2014	1	Yes	2 W	6	Daily 20 µg	No	AB (amoxicillin/clavulanate OA 875/125 mg /8H for 15D) + CLH (each 8H)
Neuprez A et al. 2014	1	Yes	At the time of diagnosis, treatment is suspended and TPTD is initiated	4	Daily 20 µg	No	-
Kim KM et al. 2014	15	N/R (15)	N/R (15)	6 (15)	Daily 20 µg (15)	No (15)	N/R (15)
Yoshiga D. et al. 2013	2	Yes (2)	1 Y (1); 2 Y (1)	3 (2)	Weekly 56.5 µg (1); daily 20 µg (1)	Yes (2)	AB (N/R) + MW + LIR
Ohbayashi Y et al. 2013	1	Yes	At the time of diagnosis, treatment is suspended and TPTD is initiated	6	Daily 20 µg	No	AB (N/R) + CLH + LIR
Kwon YD et al. 2012	6	Yes (6)	N/R (6)	1–6 (6)	Daily 20 µg (6)	Yes (6)	-
Iwamoto J et al. 2012	1	Yes	2.5 M	3	Daily 20 µg	No	AB (Amoxicillin 750 mg/D)
Choi et al. 2024	3	Yes (3)	At the time of diagnosis, treatment is suspended and TPTD is initiated (2) before** (1)	6 (3)	Daily 20 µg (3)	No (3)	AB + CLH + MW + LIR + AI
Kim et al. 2024	2	Yes (2)	N/R (2)	6 (2)	Daily 20 µg (2)	No (2)	AB (N/R) + CLH + MW + LIR + AI
Singh et al. 2025	1	Yes	2 M	6	Daily 20 µg	No	AB (N/R) + CLH + MW + LIR
Park JH et al. 2025	Conservative (8) Surgical (27)	Conservative: yes (8) Surgical: Yes (27)	Conservative: N/R (8) Surgical: N/R (27)	Conservative: > 1 year (8) Surgical: > 1 Year (27)	Conservative: daily 20 µg (8) Surgical: daily 20 µg (27)	Conservative: no (8) Surgical: no (27)	Conservative: AB + CLH + LIR + MW
Cheung et al. 2010	1	N/R	12 M	2	Daily 20 µg	No	AB
Narongroeknawin et al. 2010	1	Yes	N/R	4	Daily 20 µg subcutaneous	No	CLH (chlorhexidine rinse)
Lee et al. 2011	1	No	N/R	6	Daily 20 µg	No	AB (N/R) + CLH
Lau AN & Adachi JD. 2009	1	Yes	5–8 M	18	Daily 20 µg subcutaneous	No	AB
Tsai KY et al. 2010	1	N/R	5 M	10	Daily 20 µg	No	AB
Total	201	Yes (134); No (28); N/R (39)	Variable; N/R in most cases	1–6 M (120); 7–12 M (10); > 1 Y (64); N/R (7)	Daily (169); weekly (32)	-	AB (77); conservative (89)

Author/year of publication		Therapy during or after PTH treatment		Additional therapies	Result (improvement and healing)	Follow-up time	NICE
	Surgical	Conservative	Surgical
Yoshiga D et al. 2021	-		SCT	-	Healing (15)	3 M (15)	2-
Kim KM et al. 2023			SCT	BMP-2 Use	Healing (29) Daily: mean time: 8.35 W (17) - Weekly: Mean Time: 9.17 W (12)	6 M (29)	2 +
Mariño RQ et al. 2015	-	-	-	-	Healing in 6 W	1 M	3
Mohamed MMS et al. 2022	-	-	-	-	Healing in 2 M	2 M	3
Sim IW et al. 2020	-	AB OA (N/R), MW + DBD		Ca; Vit D	Healing (7); improvement (4); No change (4)	13 M (15)	1 +  +
Morishita K et al. 2020	-	AB (13)	SCT (7 during treatment, 6 after treatment)		Healing (19); improvement (3); no change (6); worsening(1)	Mean: 14 M (29*)	3
Javelot MJ et al. 2020	-	-	-	LLLT: - Twice a week for the first month - Twice a week for the next 6 months - Once a month until mucosal healing	Healing	26 M	3
Ohbayashi Y et al. 2020	-	AB (1- 18 W, depending on severity). Acute IV; Chronic (> 6 M): AB adjustment with antibiogram	-	-	Healing (8); improvement (4)	6 M (12)	1 +
Kim JY et al. 2019	-	-	-	-	Healing	2 M	3
Pispati A et al. 2018	SCT + Saucerization	-	-	Vitamin D3	Healing	24 M	3
Mizohata K et al. 2018	-	-	-	-	Healing	4 M	3
Jung J et al. 2017	SCT	AB (N/R)	-	BMP-2 Use	Healing (6)	N/R (6)	2-
Zushi Y et al. 2017	Implant removal	Benzalkonium chloride	SCT	-	Healing	10 M	3
Yao M et al. 2016	-	AB (amoxicillin 750 mg/D for 7 M)	SCT (6 M)	-	Healing**	2 M	3
Kakehashi H et al. 2015	-	AB (N/R) + CLH (8 patients) + CTG (2 patients)	SCT (5 patients between 0 a 7 M)	-	Healing (7) No change (1)	4–6 months—healing (1) 7–12 months—healing (3) 1–2 years—No change (1); healing (2) More than 2 years—Healing (1)	3
Yamachika E et al. 2015	-	AB (amoxicillin 750 mg for 3 M)	-	-	Healing	3 M	3
Doh RM et al. 2015	-	-	SCT (4 M post-TPTD) + implant removal	-	Healing	9 M	3
Pelaz A et al. 2014		AB ( amoxicillin/clavulanate 4 g/ for 15D	-	-	Healing (1); no change (2); worsening (1)	4–6 M- healing (1); No change (1) 7–12 M-No change (1); worsening (1)	2-
García B et al. 2014	-	-	-	-	Healing	6 M	3
Neuprez A et al. 2014	-	AB	CTG	Ca + VitD	Healing	3 M	3
Kim KM et al. 2014	-	-	SCT (12)	-	Healing (5); improvement (10)	6 M (15)	2-
Yoshiga D. et al. 2013	-	-	-	-	Healing (2)	3 M (2)	3
Ohbayashi Y et al. 2013	-	AB (clarithromycin 400 mg for 4 M) + CLH + LIR (saline solution with iodine 7%)	SCT at 4 months	-	Healing	6 M	3
Kwon YD et al. 2012	-	AB (N/R); spontaneous improvement in a patient without additional treatment	SCT (5)	-	Healing (6**)	1–3 M (6)	3
Iwamoto J et al. 2012	-	AB (piperacillin 1 g) + LIR (Saline Solution) for 5 Days	-	-	Healing	2 M	3
Choi et al. 2024	-	AB + CLH + MW + LIR + AI	-	-	Healing (3)	N/R (3)	3
Kim et al. 2024	SCT + DBD	CLH + MW + LIR	-	Ca; Vit D	Healing: 2	4–6 M (2)	3
Singh et al. 2025	-	AB (amoxicillin) + AI (ibuprofen–acetaminophen) + CLH + MW	-	-	Healing	1–2 Y	3
Park JH et al. 2025	Surgical: SCT + DBD + CTG	Conservative: AB + CLH + MW + LIR	Surgical: SCT + DBD	Conservative: Vit D Surgical: Vit D	Conservative: healing (8) Surgical: healing (27)	Conservative: N/R (8) Surgical: N/R (27)	2 +
Cheung et al. 2010	DBD	-	-	-	Healing	2 M	3
Narongroeknawin et al. 2010	DBD	CLH	-	-	Healing	4 M	3
Lee et al. 2011	SCT + DBD	-	-	-	Improvement	1 Y	3
Lau AN & Adachi JD. 2009	DBD	AB	DBD	-	Improvement	18 M	3
Tsai KY et al. 2010	SCT	-	-	-	Healing	10 M	3
Total	SCT (39); DBD (4); IM (1)	AB (84); conservative (39)	SCT (105); CTG (1)	-	Healing (163); improvement (24); No change (12); dorsening (2)	Variable; see study-level rows

AB antibiotic, CLH chlorhexidine rinses, AI anti-inflammatory, SCT sequestrectomy, LIR local irrigation, DBD debridement, PT non-surgical periodontal therapy, OA oral administration, MW mouthwashes, CTG alveolar curettage, Ca calcium, Vit D vitamin D, LLLT low level laser therapy, Y years, W weeks, D days

^*Patients discontinued or withdrew from the study after receiving PTH; clinical outcomes were reported for these patients and were included when available

^**PTH was initiated for a vertebral fracture before MRONJ was diagnosed

These publications comprised one placebo-controlled randomized clinical trial, one randomized comparative pilot clinical trial, two nonrandomized clinical trials, three cohort studies, one case-control study, five case series, eighteen case reports, and three case reports or letters/correspondence with original extractable clinical data, reflecting the heterogeneous nature of the available evidence.

Clinical outcomes were summarized descriptively from author-reported post-treatment evolution and should not be interpreted as a pooled estimate of treatment efficacy, particularly given the frequent use of concomitant conservative and surgical interventions.

Patient characteristics

Patients ranged in age from 56 to 91 years. Women accounted for 88.4% of reported cases, consistent with the epidemiology of osteoporosis.

The indication for antiresorptive therapy was osteoporosis in 156 patients and oncologic conditions in 14 patients, including multiple myeloma in 10 patients and solid tumors in 4 patients; the indication was not reported in 37 patients. Although osteoporosis-related antiresorptive exposure predominated, oncology-related cases represented a clinically relevant minority. In these patients, antiresorptive therapy may have been prescribed either for prevention of skeletal-related events in metastatic disease or for cancer treatment–induced osteoporosis; however, metastatic status, treatment intent, disease activity, and dosing context were inconsistently reported, limiting oncology-specific interpretation. In a minority of cases, MRONJ was reported in association with non-antiresorptive systemic therapies, but all included studies met the predefined criterion of teriparatide administration as part of MRONJ management.

Systemic and behavioral risk factors were also extracted when available and are summarized in Table 1. Reported factors included glucocorticoid or corticosteroid use, cigarette smoking, alcohol consumption, diabetes mellitus, rheumatoid disease or rheumatoid arthritis, cardiovascular disease, arterial hypertension, osteoarthritis, and other comorbidities or concomitant medications. Because these variables were inconsistently reported across studies, they were used to characterize the clinical context of the included patients and were not included in formal subgroup analyses.

MRONJ severity

MRONJ staging was heterogeneously reported across studies. Stage 2 and Stage 3 were the most frequently reported categories, indicating that moderate to advanced disease constituted the majority of stageable cases. Because several studies used aggregated, nonstandardized, or not-established categories, pooled stage percentages were not calculated across the entire cohort.

Anatomic distribution

Lesion site was reported for 182 patients.

Within this subset:

• Mandible only: 128 cases (70.3%)

• Maxilla only: 46 cases (25.3%)

• Both jaws: 8 cases (4.4%)

Mandibular involvement predominated among reported cases.

Prior antiresorptive exposure

Exposure duration was reported in both categorical and non-uniform formats across studies. Because several reports provided only study-level averages or ranges, individual-level pooled categorization was not consistently possible. Table 1 summarizes all extracted exposure information descriptively, whereas proportion-based analyses in the text were restricted to directly classifiable patient-level data.

Categorical exposure time was available for 201 patients:

• < 12 months: 11 patients (5.5%)

• 1–4 years: 100 patients (49.8%)

•  > 4 years: 90 patients (44.7%)

Most reported cases (94.5%) involved exposure durations of one year or longer.

The reported duration of antiresorptive exposure before MRONJ development ranged from 4 months to 17 years. Because exposure duration was reported heterogeneously across studies, further pooled quantitative analysis was not feasible.

Type of antiresorptive exposure

Bisphosphonate exposure predominated among the included reports. As categorized in Table 1, 159 patients had bisphosphonate exposure, including alendronate in 80 patients, risedronate in 30, ibandronate in 19, zoledronate in 12, minodronate in 8, pamidronate in 2, and mixed bisphosphonate exposure in 8. Denosumab-associated MRONJ was reported in 8 patients, while the antiresorptive agent was not reported in 40 patients. Because some cases involved mixed or sequential exposure and outcomes were not consistently reported at the individual-drug level, teriparatide response was not formally compared between bisphosphonate- and denosumab-associated MRONJ. Clinical improvement was reported in both contexts, but the available data do not support conclusions regarding differential effectiveness by antiresorptive type.

Teriparatide regimen

Teriparatide treatment regimen was specified in all 201 patients with evaluable clinical outcomes. Of these, 169 received daily teriparatide, typically at a dose of 20 µg, while 32 received weekly teriparatide at 56.5 µg. Overall, the available literature was predominantly based on daily dosing regimens.

Treatment duration ranged from 1 to 24 months, with short-term administration (1–6 months) most frequently described (120 patients).

Teriparatide discontinuation and adverse events

Teriparatide discontinuation or dropout after treatment initiation was reported in 13 patients. The reasons were not uniformly specified across studies. In some cases, discontinuation was attributed to adverse events, including nausea, malaise, arthralgia, facial or lower-limb edema, nausea or vomiting, and knee arthralgia. In other cases, the reason was not established or not reported. Therefore, discontinuation was recorded descriptively and was not interpreted as treatment failure unless explicitly stated in the original report. Detailed adverse events and dropout information are provided in Supplementary Table S1.

Antiresorptive interruption and concomitant management

Antiresorptive interruption status at the time of teriparatide initiation was explicitly reported for 162 patients. Among these, antiresorptive therapy was halted in 134 patients (82.7%) and was not halted in 28 patients (17.3%). In 39 patients, interruption status was not reported. Antiresorptive discontinuation was recorded only when explicitly stated in the original report; when discontinuation was not clearly documented, interruption was not inferred. The reasons for maintaining or interrupting antiresorptive therapy were rarely specified; therefore, these data should be interpreted as reported interruption status rather than as a standardized treatment strategy or evidence of a uniform clinical rationale.

Concomitant surgical or conservative interventions (e.g., debridement, sequestrectomy, and antibiotic therapy) were described in multiple reports; however, the extent and timing of these co-interventions were inconsistently quantified at the individual-patient level.

Clinical outcomes

Among the 201 patients with evaluable outcomes, complete coverage/healing was reported in 163 cases, partial improvement in 24, no change in 12, and worsening in 2. Accordingly, healing or partial improvement was described in 187 of 201 patients (93.0%). These outcomes should be interpreted descriptively, as teriparatide was frequently administered alongside other therapies and outcome definitions were harmonized from heterogeneous reports.

When reported, clinical improvement commonly appeared within the first six months of teriparatide therapy, although timing of response varied across studies.

Quality of evidence and risk of bias

The methodological quality of the included publications was heterogeneous. The evidence base was dominated by case reports and letters/correspondence with original patient-level clinical data, together with a smaller number of randomized clinical trials, non-randomized clinical trials, cohort studies, a case-control study, and case series.

According to the NICE framework for therapy studies, one placebo-controlled randomized trial was classified as level 1++, whereas the second randomized study was categorized as level 1+ due to methodological limitations. The remaining evidence largely consisted of case reports, letters/correspondence with original clinical data, and small observational studies, which inherently limit external validity.

Blinding, standardized outcome definitions, and long-term follow-up were inconsistently reported across studies. Additionally, heterogeneity in staging criteria reporting, antiresorptive exposure documentation, and adjunctive interventions further limits comparability.

Overall, while the consistency of reported clinical improvement is notable, the strength of the available evidence remains limited by study design, sample size, and reporting variability.

Discussion

This structured descriptive synthesis found that published reports frequently describe healing or partial improvement after teriparatide use in MRONJ. In the dataset, healing or partial improvement was reported in 187 of 201 evaluable patients (93.0%). However, these observations arise from a heterogeneous body of evidence dominated by case reports, small observational studies, letters or correspondence with original clinical data, and multimodal treatment contexts. Therefore, they should be interpreted as a descriptive pattern rather than evidence of therapeutic efficacy.

The consistency of reported healing or partial improvement across diverse study designs, dosing regimens, and interruption strategies is a central finding. Although derived largely from observational evidence, the convergence of outcomes across independent reports suggests a consistent observable pattern; however, causality cannot be inferred.

Because most evaluable patients received daily teriparatide at 20 µg (169 of 201; 84.1%), the apparent clinical pattern observed in this review is driven mainly by evidence derived from daily rather than weekly schedules.

Biological interpretation: remodeling restoration as a therapeutic axis

The following interpretation should be understood as a biologically reasoned framework for discussion rather than as a mechanism demonstrated by the included clinical evidence.

MRONJ is widely understood as a disorder linked to impaired bone remodeling, angiogenesis, and local immune modulation. Prolonged antiresorptive therapy suppresses osteoclastic activity, reducing remodeling turnover and potentially impairing microdamage repair. In this context, the jaws, particularly the dentoalveolar regions of the maxilla and mandible, are characterized by high physiological turnover and continuous microbial challenge and may represent uniquely vulnerable skeletal sites.

Teriparatide exerts anabolic skeletal effects by stimulating osteoblastic activity while also indirectly increasing osteoclastic activation in a coupled manner. Rather than directly eliminating necrotic bone, restoration of physiological remodeling dynamics may facilitate the resorption of damaged or non-viable bone at the lesion margins. This process could contribute to clearer biological demarcation of necrotic areas and potentially improve the conditions for surgical management. This mechanism is consistent with histomorphometric evidence from systemic skeletal sites demonstrating reactivation of coupled remodeling under teriparatide therapy.

The clinical improvements reported in MRONJ cases may therefore be interpreted within the broader framework of remodeling reactivation rather than as a purely local pharmacologic effect. These findings align with the well-established systemic skeletal actions of teriparatide in osteoporosis and suggest that restoration of remodeling may represent a biologically coherent therapeutic strategy in MRONJ.

Importantly, this interpretation does not imply direct extrapolation from osteoporosis treatment paradigms, as teriparatide in these reports was administered specifically for MRONJ and often for limited durations.

Oncologic context and therapeutic complexity

The inclusion of patients with oncologic indications adds complexity to interpretation. Antiresorptive therapy in oncology may reflect either management of bone metastases or treatment of cancer-related osteoporosis, and this distinction was not consistently reported.

Historically, teriparatide has been approached with caution in patients with malignancy due to theoretical concerns regarding osteosarcoma risk and potential stimulation of skeletal tumor activity. However, one randomized clinical trial included exclusively oncologic patients and did not report adverse oncologic progression attributable to teriparatide during the study period.

While these observations are reassuring, the available data remain insufficient to establish long-term safety in this population. Therefore, teriparatide use in oncologic MRONJ should continue to be individualized and carefully evaluated.

Because oncology-related cases constituted a minority of the extracted population and were incompletely characterized with respect to metastatic status and antiresorptive indication, conclusions specific to oncologic MRONJ should be considered exploratory.

Antiresorptive interruption and clinical variability

Antiresorptive interruption status was explicitly reported for 162 patients. In this subset, antiresorptive therapy was halted in 134 patients and was not halted in 28; interruption status was not reported in 39. Thus, antiresorptive continuation was observed in a minority of the overall dataset and should not be interpreted as half of the entire study population. Discontinuation was recorded only when explicitly stated in the original report, and interruption was not inferred when documentation was unclear.

The reasons for not interrupting antiresorptive therapy were rarely described. Continuation may have reflected different clinical scenarios, including persistent fragility-fracture risk, oncologic skeletal indications, uncertainty regarding the benefit of drug holidays, or separate decision-making between local maxillofacial management and systemic osteoporosis or oncology care. In many reports, teriparatide was presented primarily as treatment for a local MRONJ lesion, whereas the broader skeletal indication for antiresorptive therapy and the rationale for continuation or suspension were incompletely documented.

This distinction is clinically relevant because MRONJ management involves both local jawbone healing and systemic skeletal protection. Therefore, the available evidence does not allow conclusions regarding why antiresorptives were continued in some cases or whether continuation modified the response to teriparatide.

Antiresorptive type: bisphosphonates and denosumab

The type of antiresorptive therapy is clinically relevant because bisphosphonates and denosumab differ in skeletal retention, reversibility, and remodeling suppression. In this review, most cases involved bisphosphonate exposure, whereas denosumab-associated MRONJ was represented by a smaller number of patients and some cases involved mixed or sequential exposure. Although healing or improvement was reported in both contexts, the available evidence does not allow determination of whether teriparatide is more effective according to antiresorptive class. This analysis is therefore descriptive and hypothesis-generating.

Methodological considerations and risk of bias

The strength of these findings must be interpreted within the context of substantial methodological limitations. Most included studies consisted of case reports and small observational series, inherently susceptible to selection bias, reporting bias, and publication bias. Standardization of staging, exposure duration reporting, and outcome definitions were inconsistent. The inclusion criterion was defined by teriparatide administration in established MRONJ, rather than by the specific triggering medication. Therefore, etiologic heterogeneity reflects the real-world clinical spectrum of MRONJ presentations in which teriparatide has been applied. Concomitant surgical and antibiotic interventions further limit attribution of outcomes exclusively to teriparatide.

Additionally, long-term follow-up was variably reported, precluding robust conclusions regarding recurrence rates or sustained remission.

Despite these limitations, the consistency of reported clinical improvement across heterogeneous clinical contexts is noteworthy. Its recurrence across independent centers and study designs provides a signal that warrants further investigation.

An additional limitation is that variable-specific denominators differed substantially because many reports did not provide complete information across all extracted domains. Moreover, the predominance of case reports and small favorable series raises a substantial possibility of publication bias, which may inflate the apparent frequency of improvement. This also applies to systemic and behavioral MRONJ risk factors, including glucocorticoid exposure, smoking, alcohol consumption, diabetes, and inflammatory or cardiovascular comorbidities, which were tabulated when reported but could not be analyzed comparatively. Similarly, reasons for antiresorptive continuation or interruption, adverse events, and reasons for teriparatide discontinuation were inconsistently reported; therefore, these data were summarized descriptively rather than analyzed comparatively.

Implications and future directions

Taken together, these findings support the biological plausibility that restoration of physiological bone remodeling may represent a therapeutic axis in MRONJ management. Rather than viewing MRONJ solely as a localized necrotic process requiring debridement, these data suggest that systemic modulation of remodeling dynamics may complement conventional surgical approaches.

However, the current evidence base remains insufficient to establish teriparatide as standard therapy. Well-designed prospective trials with standardized staging, exposure stratification, interruption protocols, and long-term safety monitoring are required to determine its definitive role.

Conclusion

In published clinical reports, teriparatide use in MRONJ has frequently been accompanied by healing or partial improvement, including in moderate-to-advanced cases. However, the available evidence is predominantly descriptive, heterogeneous, and commonly confounded by concomitant therapies. Accordingly, the present review supports biological plausibility and clinical interest but does not establish efficacy or justify teriparatide as standard therapy.

Although limited by methodological variability and predominance of lower-level evidence, the convergence of clinical outcomes and biological plausibility suggests that remodeling restoration warrants further consideration as part of the therapeutic framework for MRONJ.

Further high-quality prospective studies are necessary to define its safety, optimal timing, duration, and patient selection criteria.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

Open Access funding provided by Colombia Consortium.

Data Availability

This article is a review of previously published studies. No new datasets were generated or analyzed during the current study; therefore, data sharing is not applicable.

Declarations

Competing interests

The authors declare no competing interests.

Editorial role disclosure

None of the authors has an editorial role affiliation with the journal to declare.