From depression to bone health: examining the skeletal effects of classical and rapid-acting antidepressants

Abstract

Depression is a brain disorder with systemic consequences including effects on bone health. Clinical and preclinical evidence has associated major depressive disorder (MDD) with reduced bone mineral density (BMD), increased bone resorption, and higher fracture risk, particularly in older adults and postmenopausal women. Plausible mediators include chronic stress, hypothalamic–pituitary–adrenal axis dysregulation, pro-inflammatory cytokines, and altered serotonergic signaling, each of which can affect bone remodeling. Classical antidepressants, particularly selective serotonin reuptake inhibitors, have been associated with adverse skeletal outcomes in human studies, while in vitro evidence indicates effects on osteoblast differentiation, mineralization, and osteoclast activity. Human biomarker studies and preclinical evidence suggest that ketamine-related compounds can modulate bone-related pathways, although these effects appear context-, dose-, exposure-, and model-dependent. Acute and repeated ketamine administration in patients with depression has been associated with changes in bone-turnover markers, including osteoprotegerin (OPG)/receptor activator of nuclear factor kappa-B ligand (RANKL)-related, while rodent studies suggest that arketamine may attenuate stress- and ovariectomy-induced reductions in BMD and normalize the OPG/RANKL ratio. Conversely, preclinical bone-regeneration and explant data indicate that ketamine can reduce bone-cell viability and impair bone healing under specific experimental conditions. The skeletal effects of ketamine, its enantiomers, and its metabolites, including (2R,6R)-hydroxynorketamine, remain preliminary and is limited to human biomarker studies and preclinical rodent/in vitro models; no clinical longitudinal BMD, bone-microarchitecture, or fracture-outcome data are currently available. This review summarizes the evidence linking depression, classical, and rapid-acting antidepressant use and bone health, and identifies gaps and priorities for future investigation, including skeletal monitoring and physiotherapy-informed rehabilitation as adjunctive clinical considerations.

Highlights  

• Major depressive disorder is linked to poorer skeletal health through stress, inflammatory, and autonomic pathways.

• It remains difficult to separate selective serotonin reuptake inhibitor–related bone risks remain difficult to separate from depression severity, frailty, falls risk, and lifestyle confounding factors.

• Ketamine-related skeletal evidence remains preliminary, with no longitudinal clinical bone mineral density or fracture-outcome data.

Graphical Abstract

Graphical Abstract.

Summations

1. Major depressive disorder is associated with poorer skeletal health through converging neuroendocrine, inflammatory, autonomic, serotonergic, and lifestyle-related pathways.

2. Classical antidepressants, particularly selective serotonin reuptake inhibitors, have been associated with lower bone mineral density and increased fracture risk in observational studies, although these associations remain difficult to separate from depression severity, frailty, falls risk, and other confounders.

3. Ketamine, its enantiomers, and metabolites may modulate bone-related biomarkers and preclinical skeletal outcomes, but current evidence remains preliminary and does not support clinical claims of skeletal protection.

Considerations

1. Much of the human evidence is observational, with incomplete adjustment for confounding by indication, depression severity, lifestyle factors, comorbidities, frailty, and falls risk.

2. Evidence on ketamine-related skeletal outcomes is limited to human biomarker studies and preclinical rodent or in vitro models, with no longitudinal clinical bone mineral density, bone microarchitecture, or fracture-outcome data.

3. Existing studies remain heterogeneous in design, population, antidepressant exposure definition, and skeletal endpoint, with limited sex-, gender-, and age-inclusive data.

Introduction

Major depressive disorder (MDD) is one of the most common mental disorders, affecting approximately 5% of adults worldwide, with women showing a higher prevalence than men (about 1.5- to 2-fold, depending on definition and population).^1–6 Depressive symptoms include persistent sadness, poor concentration, anhedonia, and physical changes such as altered appetite, sleep, and body weight^7,8; MDD is also associated with a heightened risk of suicide and is ranked as the second leading cause of disability worldwide.^9,10 The annual economic burden of MDD was increased from $236.6 billion to $326.2 billion from 2010 to 2020 in the United States alone.¹¹ Antidepressant prescribing in several populations has also increased since the COVID-19 pandemic.¹² Evidence indicates that depression could affect skeletal health and may be linked to bone loss and fracture development via direct biological processes or indirectly through different behavioural processes.¹³

Bone disease includes osteoporosis, a systemic skeletal disease that is defined by reduced bone mass and bone mineral density (BMD), degraded bone microarchitecture, and susceptibility to fragility fractures.^14,15 Osteoporosis is most common in women but also occurs in men, and it affects an estimated 200 million people worldwide.¹⁶ Osteoporosis also imposes a significant economic burden, with fracture-related costs being the most critical cost factor.^16,17 These costs further increase when indirect consequences are considered, including pain, loss of productivity, loss of independence, and psychological effects after fracture.

Although depression and osteoporosis are distinct conditions, they are interlinked. Depression is associated with reduced activity, exercise, and mechanical loading, which may indirectly drive loss of bone strength.^18,19 Conversely, individuals with impaired bone health may be susceptible to anxiety, fear of falling or fracture, and low quality of life that could exacerbate depressive symptom features.²⁰ Most of the literature reports a negative association between depression and BMD or fracture risk, although the relationship has yielded inconsistent results across cohorts.^19,21–23 In parallel, antidepressant exposure, especially long-term selective serotonin reuptake inhibitor (SSRI) use, has been linked to a significantly lower BMD and increased fracture risk in certain observational studies, but confounding by indication and incomplete adjustment for lifestyle and comorbidity are significant limitations.^2,24,25

The purpose of the present narrative review is to examine the existing literature on depression, antidepressant exposure, and bone health, with particular attention to the distinction between depression-related skeletal changes and antidepressant-associated outcomes. By synthesizing clinical, animal, and in vitro evidence, the review aims to provide a balanced overview of the depression–antidepressant–bone axis, to identify where causal inference is currently justified, and to highlight clinically relevant evidence gaps.

Review methodology

This article is a narrative review intended to synthesize and critically appraise heterogeneous clinical, animal, and in vitro literatures linking depression, antidepressant exposure, and bone health. It does not follow a formal systematic-review framework such as PRISMA, no protocol was developed or preregistered. The narrative approach was chosen because the aim was not to estimate a single pooled effect but to integrate epidemiological, mechanistic, pharmacological, and clinical-monitoring evidence that differs substantially in design, population, exposure definition, and endpoint. The reporting nevertheless follows the transparency principles expected for narrative reviews, including those reflected in the SANRA framework²⁶ and broader EQUATOR Network guidance.

Literature searches were conducted in PubMed, Web of Science, and Scopus, supplemented by Google Scholar for citation tracking and identification of additional relevant work. Searches were performed from December 2025 to June 2026, with database content considered from inception to June 23, 2026. Reference lists of key included articles and recent topical reviews were hand-searched to identify additional studies of relevance.

Search strings combined terms relating to three concepts: (1) depressive disorders, including “depression”, “major depressive disorder”, and “MDD”; (2) antidepressant pharmacotherapy, including “antidepressant”, “selective serotonin reuptake inhibitor”, “SSRI”, “serotonin–norepinephrine reuptake inhibitor”, “SNRI”, “tricyclic antidepressant”, “TCA”, “ketamine”, “(S)-ketamine”, “esketamine”, “(R)-ketamine”, “arketamine”, “(2R,6R)-hydroxynorketamine”, “HNK”, “(2R,6R)-HNK”, and “psilocybin”; and (3) skeletal outcomes and mechanisms, including “bone”, “bone mineral density”, “BMD”, “osteoporosis”, “fracture”, “bone microarchitecture”, “bone turnover markers”, “bone regeneration”, “bone healing”, “bone-cell viability”, “osteoblast”, “osteoclast”, “osteocyte”, and “bone remodelling”.

Eligible records were peer-reviewed primary research articles, including clinical observational studies, randomized controlled trials, animal studies, and in vitro work, as well as meta-analyses, systematic reviews, and authoritative narrative reviews published in English. Studies were considered for inclusion if they reported on (1) the association between depression and bone health outcomes; (2) the effects of antidepressant exposure on bone density, microarchitecture, fracture risk, or bone-cell biology; or (3) mechanistic pathways linking mood disorders and their pharmacological treatment to skeletal homeostasis. Conference abstracts, editorials, commentary pieces without supporting data, non–peer-reviewed sources, and non-English publications were excluded.

Records identified by the searches were screened by the authors at title and abstract level; full texts of potentially relevant studies were then assessed against the eligibility criteria above. Final inclusion was determined by topical relevance to the depression–antidepressant–bone axis, methodological quality of the underlying study and contribution to the mechanistic synthesis presented. Disagreements regarding inclusion or interpretation were resolved by discussion among the authors. Since this is a narrative rather than a systematic review, no formal risk-of-bias assessment was performed and no quantitative synthesis was undertaken; instead, methodological limitations of the underlying evidence base are discussed throughout the manuscript and summarized in the “Limitations and Future Directions” section.

Selective citation is a recognized limitation of narrative reviews, and several steps were taken to mitigate it. Searches were run across multiple complementary databases and supplemented by hand-searching and citation tracking rather than relying on a single source. Eligibility criteria were defined in advance and applied uniformly, without differential treatment of positive, null, or contradictory findings. Studies reporting null, neutral, or contrary results were actively sought and are explicitly discussed, including contradictory observational findings on the antidepressant–bone association, neutral animal data on fluoxetine, the (2R,6R)-HNK rodent study that did not reproduce the (R)-ketamine (arketamine) bone effect, and ketamine cytokine findings that did not correlate with antidepressant response. Where the same construct was supported by both supportive and contradictory evidence, the heterogeneity was made visible rather than smoothed over, and the underlying methodological reasons for discrepancy were discussed.

As this review spans heterogeneous clinical, animal, and in vitro literatures, a formal GRADE assessment was not undertaken. To make the strength of the underlying evidence transparent, we applied a simple 4-tier qualitative grading to each major evidence stream: strong (multiple high-quality studies with concordant findings, replication across populations, or models and adequate adjustment for known confounders); moderate (consistent overall direction but with methodological heterogeneity, incomplete confounder adjustment, or modest sample sizes); limited/emerging (small number of studies, single-center or proof-of-concept findings, or inconsistent results requiring further replication); and insufficient (too few or too heterogeneous studies to support conclusions). Gradings for human observational studies, human clinical biomarker studies, animal studies, and in vitro studies are summarized in Table 2. In line with this qualitative framework, claims of association are reported as such throughout the manuscript and are explicitly distinguished from claims of causation, the latter being reserved for findings supported by experimental intervention studies in animals or by mechanistic cellular work.

Table 2.

Strength and limitations of evidence linking depression, antidepressant exposure, and bone health across study types

Evidence stream	References	Key methodological features	Strength of evidence	Principal limitations/unaddressed confounders
Human observational (depression-BMD/fracture associations; antidepressant exposure-BMD/fracture associations)	19 ,  25,  112-117,  146,  147	Predominantly cross-sectional or retrospective cohort designs; some prospective cohorts; 1 meta-analysis and 1 systematic review of antidepressant exposure.	Moderate—direction of association is reasonably consistent, but heterogeneity and incomplete adjustment limit causal inference.	Variable depression ascertainment; inconsistent adjustment for depression severity/chronicity, frailty/falls risk, smoking, alcohol, BMI, vitamin D/calcium status, glucocorticoid exposure, menopausal status, and comorbidity; confounding by indication; underrepresentation of men and gender-diverse populations.
Human clinical biomarker (cortisol, ACTH, prolactin, OPG/RANKL, cytokines, microRNA in depression/antidepressant studies)	67 ,  128,  129,  133-135,  154,  155	Small single-center samples; mostly cross-sectional, single-infusion, or open-label repeated-infusion designs; limited replication.	Limited/emerging.	Small sample sizes; preanalytical biomarker variability; absence of longitudinal BMD, bone-microarchitecture, and fracture endpoints; biomarker changes not consistently linked to clinical bone outcomes; no validated prognostic thresholds; limited ability to isolate ketamine effects from depression improvement, concomitant medications, sex hormones, or regression to the mean.
Animal studies (stress/depression models and antidepressant-bone outcomes)	133 , 149–153, 156–158, 160,162	Diverse rodent models (CMS, CSDS, OVX); short follow-up; mostly female animals; doses not always aligned to clinical pharmacokinetics.	Moderate for stress-induced bone loss; moderate within rodent literature for arketamine bone-attenuation signals but insufficient for clinical inference.	Heterogeneous models with variable translational validity; SSRI findings inconsistent; underrepresentation of male animals; clinically irrelevant dosing in some studies; anesthetic co-exposure and bone-regeneration models may not translate directly to antidepressant-dose ketamine treatment.
In vitro studies (osteoblast/osteoclast responses to monoamines and antidepressant drugs)	147-151 ,  161,  162	Primary mouse/human osteoblast and osteoclast cultures; murine bone-marrow cultures.	Moderate for mechanistic plausibility of peripheral serotonergic effects on bone-cell function.	Frequent use of supratherapeutic drug concentrations; limited testing of ketamine enantiomers and (2R,6R)-HNK at clinically relevant concentrations; no in vitro work on esketamine at clinically used dosing; bone-explant viability findings may not translate directly to antidepressant-dose exposure in humans.

Abbreviations: ACTH, adrenocorticotropic hormone; BMD, bone mineral density; BMI, body mass index; CMS, chronic mild stress; CSDS, chronic social defeat stress; HNK, hydroxynorketamine; OPG, osteoprotegerin; OVX, ovariectomy; RANKL, receptor activator of NF-κB ligand; SSRI, selective serotonin reuptake inhibitor.

Depression and antidepressant treatments: an overview

Depression, and more specifically MDD, has a long historical lineage. The Greek term melancholia was used by Hippocrates to describe states of pathological low mood.²⁷ In 19th- and early 20th-century psychiatry, melancholia was gradually separated from other forms of mental disorder and the term depression became increasingly established in clinical nosology.^28,29 Depression can refer both to a symptom dimension, ranging from sadness to severe low mood, and to a clinical syndrome. According to the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), a major depressive episode is diagnosed when symptoms persist for at least 2 weeks and include, among other symptoms, depressed mood or anhedonia, weight or appetite change, insomnia or hypersomnia, fatigue, excessive guilt, and reduced ability to think or concentrate.⁸

Depression has been traditionally linked to altered serotonin (5-hydroxytryptamine; 5-HT) signaling when 5 patients with hypertension treated with reserpine, which depletes monoamines including 5-HT, were observed to develop depressive symptoms.³⁰ Subsequent studies also supported a link between reserpine exposure and depressive symptoms.^31,32 In parallel, iproniazid and isoniazid, initially used as tuberculosis treatments, were found to inhibit monoamine oxidase (MAO) and improve mood.^33,34 These observations contributed to the monoaminergic hypothesis of depression, which proposed that reduced monoamine signaling contributes to depressive pathophysiology.

Monoamine-based antidepressant medications

The first commercially available antidepressant, imipramine, was introduced in 1958, ending an era in which electroconvulsive therapy and psychotherapy were the principal available treatments for depression.³⁵ Early antidepressant development was guided by the monoamine hypothesis, which proposed that depression arises partly from reduced synaptic availability of monoamine neurotransmitters such as 5-HT and noradrenaline.^36,37 Tricyclic antidepressants (TCAs) were among the first drugs to target this mechanism; tertiary amine TCAs (eg, imipramine and amitriptyline) preferentially block 5-HT reuptake, whereas secondary amines (eg, desipramine and nortriptyline) more strongly inhibit noradrenaline transporters.² Their clinical utility is, however, limited by adverse effects, including seizures, cognitive impairment, and lethality in overdose, attributable to broad off-target receptor activity.³⁸

To improve safety and specificity, SSRIs were developed to selectively block the serotonin transporter, thereby increasing extracellular 5-HT and promoting adaptive neuroplastic changes that may underline their therapeutic effects.³⁹ The first SSRI, zimeldine, was approved in Sweden in 1982 but withdrawn a year later, owing to an increased incidence of Guillain–Barré syndrome, an autoimmune disorder of peripheral nerves.⁴⁰ The clinical breakthrough came in late 1987 with the US Food and Drug Administration approval of fluoxetine (Prozac), which rapidly became one of the most widely prescribed antidepressants worldwide and is widely regarded as having ushered in the modern era of antidepressant treatment.⁴¹ The promising results of fluoxetine were followed by the introduction of additional SSRIs, including sertraline, paroxetine, fluvoxamine, citalopram, and escitalopram, which differ in pharmacokinetic properties and tolerability but share a common mechanism of enhanced serotonergic signaling.⁴¹

Serotonin-norepinephrine reuptake inhibitors (SNRIs) were then developed to engage both serotonergic and noradrenergic mechanisms.^42–44 Unlike SSRIs, SNRIs inhibit both 5-HT and noradrenaline reuptake, and some evidence suggests that this broader monoaminergic profile can produce faster antidepressant effects in some contexts.^43,45,46 Venlafaxine, the first widely used SNRI, was approved for treatment of depression in the early 1990s.⁴⁷

In addition to their primary action on monoamine reuptake, classical antidepressants exert downstream molecular, neuroplastic, and immunological effects that are increasingly recognized as integral to their therapeutic profile. Chronic SSRI, SNRI, and TCA treatment can enhance brain-derived neurotrophic factor (BDNF) signaling, support adult hippocampal neurogenesis, and promote structural and functional synaptic plasticity.⁴⁸ These agents may also modulate hypothalamic–pituitary–adrenal (HPA) axis activity and reduce circulating levels of pro-inflammatory cytokines, including tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β), although the magnitude and consistency of these effects vary by study design and treatment response.^49,50 Importantly, several of these pathways—peripheral serotonergic signaling, glucocorticoid exposure, and cytokine activity—directly intersect with mechanisms that regulate osteoblast and osteoclast function, providing a biological framework for the skeletal associations examined in the “Links between Depression and Bone Health” section.

Rapid-acting antidepressants

Beyond the monoamine-targeting agents discussed above, a newer generation of rapid-acting antidepressants has emerged over the past 2 decades. The most extensively investigated is ketamine, but classical psychedelics—most notably psilocybin and, to a lesser extent, lysergic acid diethylamide and ayahuasca—have also produced rapid and durable antidepressant effects in clinical trials, primarily through 5-HT2A receptor–mediated mechanisms coupled to downstream neuroplastic adaptations.⁵¹ Robust preclinical and clinical evidence on skeletal effects currently exists only for ketamine and its enantiomers and metabolites; the bone-related consequences of psychedelic-assisted therapy remain largely unexplored and represent a clear gap for future investigation. The remainder of this section therefore focuses on ketamine and its derivatives.

(R,S)-ketamine, commonly referred to as ketamine, is a noncompetitive N-methyl-D-aspartate (NMDA) receptor antagonist originally introduced as an anaesthetic.^52,53 It has gained recognition as a rapid-acting antidepressant because sub-anesthetic doses can reduce depressive symptoms within hours in some patients.^54–58 Early evidence of ketamine’s psychiatric use dates back to Khorramzadeh and Lotfy,⁵⁹ who reported that intravenous ketamine facilitated psychotherapy among psychiatric patients. The first placebo-controlled study highlighting ketamine’s antidepressant potential was published in 2000.⁵⁷ Subsequent clinical trials and meta-analyses indicate efficacy in unipolar and bipolar depression, particularly in treatment-resistant populations, and ketamine has also been associated with rapid reductions in suicidal ideation.^54,58,60,61

The mechanism by which ketamine exerts rapid antidepressant effects has been hypothesized to involve NMDA receptor modulation, increased glutamate release, and activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors, which are important for synaptic plasticity and neurogenesis.^62–65 Ketamine also influences growth-factor and intracellular signaling pathways, including vascular endothelial growth factor, insulin-like growth factor-1, BDNF, and mammalian target of rapamycin (mTOR), thereby contributing to synaptic strengthening.⁶⁴ In parallel with these central neuroplastic actions, acute ketamine administration has been reported to modulate inflammatory activity and to shift inflammatory bone-turnover biomarkers, including the OPG/RANKL axis, toward a less resorptive profile in patients with MDD.^66,67 These biomarker observations are hypothesis-generating and are considered with appropriate caveats regarding their biomarker-level nature, single- or repeated-infusion design, and absence of longitudinal BMD or fracture endpoints in the “Effects of Ketamine, Its Enantiomers, and Metabolites on Bone  Health” section.

Although ketamine is a promising antidepressant, racemic ketamine remains an off-label treatment for depression in many jurisdictions and requires careful monitoring. Even at sub-anesthetic doses, ketamine is associated with short-lived dissociative symptoms, perceptual changes, and potential misuse liability.^68–70 The esketamine enantiomer of ketamine, has higher NMDA receptor binding affinity and is a more potent anesthetic, but it can also produce dissociation and perceptual adverse effects.⁷¹ Intranasal esketamine is approved for treatment-resistant depression and for depressive symptoms in adults with MDD with acute suicidal ideation or behavior in some regulatory jurisdictions.^72,73 By contrast, the arketamine enantiomer has shown favorable behavioural and tolerability profiles in preclinical studies,⁷⁴ but clinical evidence remains limited: An early open-label pilot study reported encouraging antidepressant signals, whereas adequately powered controlled trials are still needed before firm conclusions can be drawn.^70,75

After administration, ketamine is rapidly metabolized into norketamine, which is further hydroxylated to form hydroxynorketamine (HNK). The primary HNK metabolites found in human and rodent plasma are (2S,6S)-HNK and (2R,6R)-HNK. Of these, (2R,6R)-HNK has been identified as a mediator of ketamine’s antidepressant-relevant effects in mouse models.^76–78 Unlike ketamine, (2R,6R)-HNK does not produce sensory dissociation or abuse-related behaviors in mice.^76,79,80 Furthermore, (2R,6R)-HNK does not inhibit NMDA receptor function at antidepressant-relevant concentrations in the same way as ketamine.^53,81 Instead, it has been shown to enhance presynaptic glutamate release, upregulate AMPA receptor expression, and engage metabotropic glutamate receptor pathways,⁸² as well as BDNF- and mTOR-related signaling in some experimental contexts.^76,83,84

Bone and bone modeling/remodeling

Bone homeostasis is tightly regulated by systemic and local factors that overlap with neuroendocrine and immune pathways implicated in mood disorders. Understanding these shared molecular mediators provides a framework for exploring how antidepressants may influence skeletal health.

Bone is a dynamic tissue that undergoes adaptation in response to mechanical, metabolic, and hormonal demands. This adaptability is achieved through 2 distinct but complementary processes: bone modeling and bone remodeling.⁸⁵ Bone modeling shapes bone during growth and alters geometry through formation of new bone by osteoblasts or resorption of bone by osteoclasts on separate bone surfaces.^86,87 This contrasts with bone remodeling, in which osteoclast and osteoblast activity occur sequentially in a coupled manner at the same location.⁸⁶ Bone remodeling, first described by Roux in 1892, can occur on periosteal, endocortical, trabecular, and intracortical bone surfaces. In homeostatic equilibrium, resorption and formation are balanced so that old bone is continuously replaced by new tissue and bone can adapt to mechanical load and strain.⁸⁸ Osteoclasts and osteoblasts closely collaborate in the basic multicellular unit (BMU).^89,90 Osteocytes are also involved by acting as mechanosensors and coordinating the requirement for bone remodeling.⁸⁵ BMU organization differs between cortical and trabecular bone, reflecting differences in microarchitecture, surface-to-volume ratio, and remodeling dynamics; trabecular bone generally exhibits more active remodeling than cortical bone because of its larger surface area.⁹⁰ These compartmental differences are not purely morphological, however, because osteoblasts derived from distinct bone types, including trabecular, cortical, and subchondral bone, can retain bone-type-related differences in behavior and osteogenic phenotype.⁹¹

The remodeling cycle consists of consecutive activation, resorption, reversal, formation, and termination phases over several weeks.⁸⁶ Activation involves detection of a remodeling signal, which can be hormonal (eg, estrogen or parathyroid hormone [PTH]) or mechanical (eg, loading or local microdamage), prompting bone-lining cell retraction and recruitment of osteoclast precursors.^89,90 During resorption, multinucleated osteoclasts degrade bone prior to apoptosis.⁸⁶ During reversal, monocytes and pre-osteoblasts prepare the bone surface for new formation and signal osteoblast differentiation.⁹² During formation, osteoblasts synthesize new type I collagen-rich matrix that gradually mineralizes.^86,90 Finally, during termination, approximately 50%-70% of osteoblasts undergo apoptosis, with the remainder becoming osteocytes or bone-lining cells.⁹²

At the molecular level, bone remodeling is orchestrated by a complex interplay of systemic hormones, growth factors, and local signaling pathways.⁹³ The receptor activator of nuclear factor kappa-B (RANK), its ligand (RANKL), and osteoprotegerin (OPG) form a central signaling system that regulates osteoclast differentiation and activity.⁹⁴ RANKL, produced by osteoblasts and osteocytes, binds to RANK on osteoclast precursors to drive maturation, whereas OPG acts as a soluble decoy receptor that neutralizes RANKL and protects against excessive resorption.⁹⁴ This balance is influenced by inflammatory mediators. Pro-inflammatory cytokines, including tumor necrosis factor-alpha (TNF-α), IL-1β, and IL-6, promote RANKL expression, and suppress OPG, thereby enhancing osteoclastogenesis.⁹⁵ In parallel, prostaglandins such as prostaglandin E2 (PGE2) and downstream activation of nuclear factor kappa-B (NF-κB) signaling further amplify bone-resorptive activity.⁹⁶

This multi-phase remodeling process provides several points at which bone formation or resorption can be regulated to control bone mass, architecture, and mineralization.⁸⁹ Analyzing the total cortical and trabecular compartments of a single bone enables predictions about how remodeling activities are coordinated to control mechanical rigidity and load-bearing strength, particularly when they are disrupted by systemic disease, inflammation, or psychiatric conditions such as depression.

Osteoporosis—the most common type of bone disease

Osteoporosis affects more than 200 million people worldwide and is especially prevalent after menopause and with advancing age.^14,16 It is also associated with a high lifetime fracture burden: approximately 1 in 3 women and 1 in 5 men over the age of 50 will experience an osteoporotic fracture.⁹⁷ Osteoporosis is characterized by low bone mass and deteriorated architecture due to an imbalance in bone remodeling, where resorption exceeds formation.^14,98 The World Health Organization has defined osteoporosis as a progressive systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture.^14,99 The immediate clinical consequence is fracture, with vertebral and hip fractures associated with substantial morbidity and mortality.^98,100,101 Several conditions can disrupt remodeling and contribute to osteoporosis, including menopause, aging, endocrine disorders, immobilization, nutritional disorders, inflammatory arthropathies, and medication exposure.^14,102

Osteoporotic fractures are preventable, and available pharmacological treatments are conventionally grouped into antiresorptive agents that inhibit osteoclasts, bone-forming (anabolic) agents that stimulate osteoblasts and dual-acting drugs that combine both actions.¹⁰¹ The principal anti-resorptive agents include: (1) denosumab, a fully human monoclonal antibody that binds RANKL and thereby prevents osteoclast differentiation, function, and survival^101,103,104; (2) bisphosphonates (eg, alendronate, risedronate, and zoledronate), which adsorb to bone mineral and, after uptake by resorbing osteoclasts, disrupt the mevalonate pathway and trigger osteoclast apoptosis^92,102,105; (3) calcitonin, a peptide hormone that suppresses osteoclast activity through calcitonin receptors expressed on these cells^106,107; and (4) selective estrogen receptor modulators (SERMs; eg, raloxifene), which act as tissue-selective estrogen receptor agonists in bone and reproduce antiresorptive estrogen effects while avoiding stimulation of breast and endometrial tissue.^101,102 Potential natural agents, such as broccoli-derived compounds, may also support bone health by modulating osteoblast and osteoclast activity.^108–110 Bone-forming therapies include the PTH analogue teriparatide and the PTH-related protein (PTHrP) analogue abaloparatide, which, when administered intermittently, activate the PTH1 receptor on osteoblasts to drive new bone formation. The dual-acting agent romosozumab is a monoclonal antibody that neutralizes sclerostin, an osteocyte-derived inhibitor of Wnt signaling, thereby simultaneously stimulating bone formation and reducing resorption.^101,111

Anti-resorptive treatments increase BMD partly by inducing osteoclast apoptosis and inhibiting bone resorption.¹⁰⁵ The increase in BMD is therefore caused by filling of the remodeling space followed by increased bone-tissue mineralization because bones are not remodeled as frequently after treatment initiation. Bone architecture is not generally fully restored, although cortical bone volume and bone strength can increase in patients treated with denosumab, a RANKL-blocking antibody.¹¹² Therefore, if a patient initially has very low bone mass, anti-resorptive treatments may not increase BMD enough to optimally prevent future fractures.

Bone-forming treatments stimulate osteoblasts and promote a positive remodeling balance. This can, in turn, stimulate osteoclasts, which may limit treatment effects, and the osteoblast response to continued stimulation can also level off over time. Patients with very low bone mass or a suboptimal response to bone-forming treatment may therefore retain low BMD even after therapy. Only a few studies have examined whether the coupling of bone resorption and formation can be overcome by combining these therapies. Romosozumab effectively uncouples bone resorption and formation by inhibiting sclerostin.¹¹¹ The remaining challenge is that although inhibition of bone resorption remains throughout the treatment period, stimulation of bone formation diminishes despite continued treatment.

Despite the availability of antiresorptive and anabolic agents, their inability to fully restore bone mass and architecture, alongside long-term safety concerns and incomplete fracture-risk reduction, highlights a need for complementary approaches. Drugs with pleiotropic effects on inflammation, oxidative stress, and neuroendocrine signaling, including antidepressants, have therefore attracted increasing attention for their potential impact on bone turnover.

Links between depression and bone health

A growing body of clinical and preclinical evidence indicates that MDD is associated with reduced BMD and increased risk of osteoporosis and fractures.^113–116 The current evidence base, however, is predominantly observational, and these associations should not be interpreted as causal without explicit consideration of substantial confounding. Important potential confounders include depression severity and chronicity, frailty and falls propensity, smoking, alcohol consumption, BMI and overall nutritional status, vitamin D and calcium intake, current or prior glucocorticoid exposure, menopausal status and gonadal hormone exposure, and the broader burden of medical comorbidity. Few primary studies adjust for all of these. Conversely, studies of depression, as an exposure rarely report bone outcomes separately for treated and untreated patients, so the “depression effect” captured in such cohorts may incorporate an unmeasured contribution from concurrent antidepressant therapy. Disentangling these sources of variance is therefore a central interpretive challenge in the human literature and is one of the reasons this review draws on complementary preclinical and in vitro evidence. Throughout the sections that follow, claims of association are distinguished from claims of causation; representative clinical, biomarker, animal and in vitro evidence is summarised in Table 1, and the strength of evidence supporting each finding is graded in Table 2.

Table 1.

Depression, antidepressant exposure, and bone health

Conditions/intervention	Population/model	Study comparators	Outcome(s)	Direction/results	Study design	Reference
Depression/anxiety and skeletal outcomes
Self-reported depression	Perimenopausal women from a community cohort	No self-reported depression	BMD (hip, spine)	Hip BMD ↓ (~7.8%); spine BMD NS	Cross-sectional/population-based	117
Depressive symptoms	Older women and men	Lower/no depressive symptoms	BMD (total hip)	Hip BMD ↓ in depressed participants; association remained significant after adjustment; strongest signal in Caucasian women	Cross-sectional/population-based	118
Depressive disorder	Adults from a national cohort	Individuals without depressive disorder	Osteoporosis incidence	Osteoporosis risk ↑; strongest signal at 35-49 years; antidepressant use associated with lower osteoporosis risk within the depressed cohort	Retrospective longitudinal cohort	113
Depressive disorder	Adults from a national cohort	Individuals without depressive disorder	Vertebral fracture risk	Vertebral fracture risk ↑; stronger signal in adults ≥50 years, females, and individuals with medical comorbidities	Retrospective longitudinal cohort	114
MDD	Adolescents	Healthy controls	BMD	Boys with MDD: BMD ↓; girls with MDD: NS	Cross-sectional clinical study	119
MDD	1842 participants with MDD and 17 401 controls across included studies	Non-depressed controls across included studies	BMD (lumbar spine, femur, hip)	Women with MDD: lumbar spine and femur BMD ↓; men with MDD: hip BMD ↓	Meta-analysis of cross-sectional studies	115
Depressive severity	Population-based adult cohort	Lower depressive severity/non-depressed comparison	BMD	Higher depressive severity associated with lower BMD	Observational population-based study	19
Anxiety level	Postmenopausal women assessed for osteoporosis	Lower anxiety	BMD; fracture-risk modifiers	Higher anxiety associated with lower BMD; effect stronger in postmenopausal women	Cross-sectional clinical study	123
Anxiety	Postmenopausal women receiving oral bisphosphonates	Lower versus higher anxiety tertiles	BMD change; treatment adherence	Higher anxiety associated with greater BMD improvement and higher bisphosphonate adherence	Two-year observational follow-up study	121
Human endocrine, inflammatory, and biomarker evidence
MDD	Premenopausal women	Healthy controls	Urinary/plasma cortisol; BMD	Cortisol ↑; BMD ↓; negative cortisol–BMD correlation	Cross-sectional clinical study	128
MDD	Premenopausal women	Healthy controls	PTH; ACTH; calcium/vitamin D; BMD	MDD associated with PTH ↑, ACTH ↑ at selected time points, calcium/vitamin D ↓, and lower or declining BMD measures over follow-up	36-month prospective clinical study	129
MDD	Adults with first-episode MDD	Non-depressed controls	Bone-turnover markers (osteocalcin, ALP, NTX/CTX); BMD	Bone-turnover markers largely unchanged despite lower BMD	Cross-sectional clinical biomarker study	134
Mental disorders/depression-related bone-health mechanisms	Adults	Healthy controls	miR-26b-5p, miR-377-3p; TNRC6B/HSPA8-related pathways	MicroRNAs identified as candidate biomarkers/pathways linking mental disorders and bone-health mechanisms	Human mechanistic/biomarker study	136
Antidepressant exposure and skeletal outcomes
Serotonergic antidepressant use, including SSRI/SNRI use	Adults in a population database	Non-current use/non-use comparison groups	Fracture risk	Current SSRI/SNRI use associated with increased fracture risk	Population-based nested case–control study	145
Antidepressant exposure	Adults across included studies	Non-users across included studies	Osteoporosis; fracture risk; BMD	Antidepressant-associated skeletal risk modified by age, sex, menopausal status, and other factors	Systematic review and meta-analysis	147
Classical antidepressant mechanistic evidence
SSRIs	Osteoblast and osteoclast cultures	Vehicle/control cultures	Osteoblast function; mineralization; osteoclast formation and resorption; ALP; apoptosis	Osteoblast function ↓; mineralization ↓; ALP ↓; apoptosis at high concentrations; osteoclast formation and resorption ↓ under some conditions	In vitro studies	148 , 149
5-HT/5-HTT manipulation	Wild-type and 5-HTT-knockout mice	Wild-type/control or pharmacological inhibition, as applicable	Bone formation; bone mass	5-HTT inhibition or deletion reduced bone formation and bone accrual; increased peripheral gut-derived 5-HT suppressed osteoblastic activity; LRP5-mediated reduction in gut 5-HT associated with increased bone formation and bone mass	Animal/mechanistic studies	150 , 151
Fluoxetine	Adult rats exposed to CMS	Vehicle-treated CMS and/or non-CMS controls	Cortical thickness; mechanical strength; trabecular indices	Fluoxetine improved cortical thickness and mechanical strength; CMS reduced trabecular indices regardless of treatment	Animal/mechanistic study	153
Fluoxetine	Mice	Vehicle-treated animals	Bone histomorphometry (trabecular bone volume, bone-formation rate)	Trabecular bone formation ↑	Animal/mechanistic study	152
Rapid-acting antidepressants and skeletal biomarkers/bone outcomes
IV ketamine, single infusion	Adults with MDD	Baseline within MDD group; healthy controls	OPG/RANKL ratio; RANKL; OPN	MDD: OPG/RANKL ↑, RANKL ↓, OPN ↑; healthy controls: NS	Clinical biomarker study	67
Repeated IV ketamine	102 patients with unipolar or bipolar depression	Baseline; post-infusion time points (day 13, day 26)	Plasma bone markers: leptin, OPG, DKK1, osteocalcin, sclerostin, OPN, PTH, FGF23	Leptin and OPG ↑ at days 13 and 26; DKK1 ↑ at day 13; osteocalcin, sclerostin, OPN, PTH, and FGF23 ↓ at days 13 and 26; sex-specific differences in leptin, osteocalcin, and sclerostin	Open-label clinical biomarker study	154
Ketamine	Rat spongious and cortical bone explants; rat calvarial bone-healing model	Ketamine-free control; single versus repeated ketamine–xylazine anesthetic exposure	Bone-cell viability; calvarial bone healing	High-concentration ketamine reduced bone-explant viability; repeated anesthetic exposure associated with impaired calvarial bone healing	In vitro and animal bone-regeneration study	162
Arketamine versus esketamine	Mice exposed to CSDS	Vehicle-treated CSDS mice	OPG/RANKL ratio	Arketamine increased OPG/RANKL ratio and this correlated with antidepressant-like behavior; esketamine NS	Animal/mechanistic study	156
Arketamine	OVX mice	Vehicle-treated OVX mice	Cortical BMD; total BMD	Arketamine ameliorated OVX-induced bone loss and restored cortical and total BMD	Animal/mechanistic study	157
Arketamine versus esketamine	OVX mice	Vehicle-treated OVX mice	Cortical BMD; total BMD	Arketamine ameliorated cortical and total BMD reduction; esketamine NS	Animal/mechanistic study	158
Arketamine; (2R,6R)-HNK	Mice exposed to CSDS	Vehicle-treated CSDS mice	OPG/RANKL ratio; bone loss	Arketamine reversed CSDS-induced bone loss and normalized OPG/RANKL ratio; (2R,6R)-HNK showed no such effect	Animal/mechanistic study	160
Stress-related mechanistic pathways relevant to depression–bone crosstalk
Glucocorticoid exposure (dexamethasone)	Neonatal mouse calvarial bone	Baseline/vehicle	RANKL; OPG; osteoclast differentiation	RANKL and OPG ↑; RANKL/OPG ratio ↑; osteoclast differentiation/resorptive signaling ↑	In vitro mechanistic study	126
Glucocorticoids	Mouse bone tissue and bone marrow stromal cell cultures	Baseline/vehicle	Osteoblastogenesis; osteoblast and osteocyte apoptosis; bone formation	Osteoblastogenesis ↓; osteoblast and osteocyte apoptosis ↑; bone formation ↓	In vivo and ex vivo mechanistic study	127
Sympathetic activity/noradrenaline	Mice	Baseline/control conditions	Trabecular bone; bone formation	Bone noradrenaline ↑ associated with bone formation ↓ and trabecular bone loss	Animal in vivo mechanistic study	133

Abbreviations: 5-HT, serotonin/5-hydroxytryptamine; 5-HTT, serotonin transporter; ACTH, adrenocorticotropic hormone; ALP, alkaline phosphatase; BMD, bone mineral density; CMS, chronic mild stress; CSDS, chronic social defeat stress; CTX, C-terminal telopeptide; DKK1, Dickkopf-related protein 1; FGF23, fibroblast growth factor 23; HNK, hydroxynorketamine; HSPA8, heat shock protein family A member 8; IV, intravenous; LRP5, low-density lipoprotein receptor-related protein 5; MDD, major depressive disorder; NS, not significant; NTX, N-terminal telopeptide; OPG, osteoprotegerin; OPN, osteopontin; OVX, ovariectomy; PTH, parathyroid hormone; RANKL, receptor activator of NF-κB ligand; SNRI, serotonin-norepinephrine reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor; TNRC6B, trinucleotide repeat containing 6B; y, years.

Studies generally report that depression is linked to lower BMD, although the extent and anatomical location of this association depend on age, sex, and ethnicity.¹¹⁶ For instance, depression has been associated with lower hip BMD in perimenopausal women (by approximately 7.8%), whereas spine BMD was not significantly different in that cohort.¹¹⁷ Conversely, an association between depression and total hip BMD was identified in older adults, with the strongest signal observed in Caucasian women.¹¹⁸ These findings highlight the importance of age, sex, ethnicity, and skeletal site as potential modifiers.

Longitudinal and population-based studies provide additional support for this association. Adults with depressive disorder have been reported to be more likely to develop osteoporosis than non-depressed individuals,¹¹³ and depressive disorder has been associated with higher vertebral fracture risk, particularly among adults aged 50 years or older, women, and individuals with medical comorbidities.¹¹⁴ Developmental stage may also matter: adolescent boys with MDD show lower BMD than age-matched controls, whereas such differences were not observed in girls in 1 cross-sectional study (Table 1).¹¹⁹ A meta-analysis of 21 cross-sectional studies, including 1842 participants with MDD and 17 401 controls, confirmed that BMD outcomes are influenced by age and sex. Women with MDD exhibited lower BMD in the lumbar spine and femur, whereas men appeared more affected at the hip.¹¹⁵ This sex difference was more pronounced in studies using clinician-administered diagnostic interviews compared with self-report measures,¹¹⁵ consistent with broader evidence that self-reported tools tend to overestimate depression prevalence and increase false positives.¹²⁰

Comorbid anxiety may further contribute to skeletal vulnerability.^121,122 Patients with higher anxiety levels have been reported to show lower BMD, although this association is moderated by age, menopausal status, and depressive symptom severity.¹²³ Greater symptom chronicity is also associated with reduced lumbar spine BMD. Additionally, higher anxiety levels correlate with lower vitamin D concentrations independently of depression in postmenopausal women, suggesting that anxiety may contribute to bone loss and reduced quality of life in aging populations.¹²⁴

Bone loss in patients with MDD has been linked to increased adrenocorticotropic hormone (ACTH) secretion, HPA-axis hyperactivation, OPG-RANKL pathway dysregulation, inflammation, and autonomic imbalance.²⁰ Psychological stress profoundly affects endocrine function, primarily through HPA-axis activation.¹²⁵ Chronic HPA-axis dysregulation leads to persistently elevated cortisol levels, which can impair bone formation and enhance bone resorption by increasing RANKL expression and shifting the RANKL/OPG balance toward resorptive signaling in osteoblast-lineage cells.¹²⁶  In vitro studies further support these effects by showing that glucocorticoids reduce osteoblastogenesis and promote apoptosis of osteoblasts and osteocytes.^125,127 Clinically, premenopausal women with MDD exhibit elevated urinary and plasma cortisol concentrations that correlate inversely with BMD.¹²⁸ Increased PTH and ACTH accompanied by reduced BMD have been reported in pre- and postmenopausal women with MDD, whereas calcium and vitamin D differences suggest that nutritional factors may further influence bone outcomes.^20,129 Stress-induced inflammation is also proposed to contribute to skeletal deterioration: mechanistic studies indicate that pro-inflammatory TNF-α, IL-1, IL-2, and IL-6 increase RANK expression, enhance RANKL signaling, and suppress OPG production, collectively promoting osteoclastogenesis. Elevated concentrations of these cytokines have been reported in adults with MDD,¹³⁰ suggesting that inflammatory activation may influence both neurocognitive and skeletal functions in depression.

Endocrine mediators also participate in this process, with prolactin emerging as a stress-responsive hormone of relevance to bone health. Anterior-pituitary secretion of prolactin can be altered under chronic psychological stress and has been reported to differ in some patients with MDD. In bone, prolactin acts on prolactin receptors expressed on osteoblast-lineage cells, where it can upregulate expression of osteoclastogenic cytokines and shift the RANKL/OPG balance toward enhanced osteoclastogenesis and increased bone resorption.¹³¹ The relevance of this pathway is reinforced by reports that antidepressants can, in some cases, alter circulating prolactin, although population-level data suggest that clinically meaningful prolactin elevation is not universal among antidepressant users.¹³² The contribution of antidepressant-associated prolactin changes to skeletal outcomes therefore remains uncertain and warrants targeted study. In parallel, sympathetic overactivity, a feature of MDD and chronic stress, raises local noradrenaline concentrations in bone, activating β-adrenergic receptors on osteoblasts and osteoclasts and thereby promoting trabecular bone loss in animal models.¹³³ Finally, gonadal hormones such as estrogen and testosterone play critical roles in maintaining bone integrity, with estrogen protecting against osteoblast apoptosis and testosterone supporting cancellous bone formation.¹²⁵

Although depression has been associated with alterations in bone metabolism, relatively few studies have examined depressive symptoms in relation to specific bone-turnover markers. Findings for osteocalcin, a bone-formation marker produced by osteoblasts, remain inconsistent, while alkaline phosphatase (ALP) and telopeptides (C- and N-terminal) often appear unchanged despite reductions in BMD among depressed individuals.^134,135 Further mechanistic insight was provided by Houtenbos et al.,¹³⁶ who suggested that microRNAs miR-26b-5p and miR-377-3p, could serve as candidate biomarkers of bone-disease treatment response in individuals with mental disorders. These microRNAs were linked to regulation of trinucleotide repeat containing 6B (TNRC6B), heat shock protein family A (Hsp70) member 8 (HSPA8), and related signaling pathways.¹³⁶ Emerging evidence also suggests that 5-HT, central to mood regulation and altered in depression, modulates bone metabolism through complex central, peripheral, and local pathways.^137,138

Other factors commonly associated with depression and poor bone health include physical inactivity, poor nutrition, smoking, and alcohol consumption. Smoking is linked to reduced estrogen levels and lower BMD,¹³⁹ while alcohol consumption can inhibit bone-cell proliferation and differentiation.¹⁴⁰ Nicotine, the principal active component of tobacco, alters hormonal balance and inhibits the differentiation and proliferation of mesenchymal stem cells into osteoblasts; however, at low concentrations, it may transiently enhance both processes.^125,141 Nicotine also promotes osteoclastogenesis by upregulating TNF-α expression, thereby increasing RANKL signaling.¹⁴² Alcohol affects bone through multiple mechanisms: It downregulates insulin-like growth factor expression; increases TNF-α, IL-1β, and IL-6; elevates sclerostin expression and inhibits Wnt signaling; and upregulates NADPH oxidase enzymes that generate reactive oxygen species and further enhance RANKL expression.^125,143

Classical antidepressants and bone health

The relationship between antidepressant use and bone health has gained increasing attention. Although antidepressants are widely prescribed to treat depression, concerns persist regarding their potential impact on bone metabolism and fracture risk, particularly with SSRIs, which have been repeatedly associated with lower BMD and increased fracture incidence in observational studies.^19,144,145 While several studies report a significant adverse association between depression and bone outcomes, others fail to confirm this association,^22,146 reflecting both genuine heterogeneity and inconsistent adjustment for the confounders detailed above.

A central methodological challenge in this literature is separating the effects of antidepressant exposure on bone from the effects of the underlying depression that prompts prescription. Observational studies that compare medicated patients with non-depressed controls, or with untreated depressed patients, may confound drug exposure with disease severity, chronicity, and treatment-seeking behavior. Three complementary lines of evidence help to disentangle the two. These include preclinical studies of antidepressants in healthy, non-stressed animals, factorial preclinical designs that cross a stress or depression-like model with antidepressant intervention, and in vitro studies on osteoblast and osteoclast cultures that isolate osteotropic drug effects from whole-organism disease states.

A recent systematic review by Hu et al.¹⁴⁷ emphasized that the link between antidepressant use and osteoporosis is modulated by age, sex, menopausal status, and genetic predisposition, indicating that the skeletal impact of antidepressants is not uniform across populations. Clinical studies such as Wadhwa et al.²⁵ link SSRI use to increased fracture rates and reduced BMD, underscoring the importance of considering BMD and fracture risk in patients receiving long-term antidepressant treatment, particularly when additional risk factors are present.²⁵

Experimental studies suggest that SSRIs may influence bone homeostasis through both central and peripheral serotonergic mechanisms. In vitro evidence indicates that SSRIs can increase osteoclast activity and suppress osteoblast function, under some experimental conditions.^148,149 Notably, both 5-HT and its transporter (5-HTT) are expressed in bone cells, and inhibition or deletion of 5-HTT in mice has been shown to reduce bone formation and bone accrual, whereas suppression of gut-derived 5-HT via low-density lipoprotein receptor-related protein 5 (LRP5) signaling increases BMD, indicating that serotonergic signaling actively regulates bone in vivo.^150,151

It has also been demonstrated that several SSRIs exerted dose-dependent inhibition of osteoclast number and bone resorption, reduced osteoblast mineralization and ALP activity, and induced apoptosis in both osteoclasts and osteoblasts at high concentrations¹⁴⁹ (Figure 1). Complementary preclinical in vivo SSRI studies map onto the 3-pillar framework introduced above. In healthy, non-stressed mice, fluoxetine administered alone increased trabecular bone formation,¹⁵² whereas in a factorial chronic mild stress × fluoxetine design in rats, the underlying stress reduced trabecular bone quality independently of fluoxetine treatment, while fluoxetine itself improved cortical thickness and mechanical strength.¹⁵³ Together with the aforementioned in vitro findings, these data suggest that the adverse SSRI-bone signals reported in some clinical cohorts may reflect, at least in part, the underlying depression and its associated lifestyle and comorbidity burden rather than a direct skeletal effect of the drug itself.

Figure 1.

Schematic overview of the depression–bone crosstalk. (A) Depression- and stress-related upstream inputs to bone remodeling. Inputs are organized into 5 main categories: (i) HPA-axis activation (cortisol); (ii) systemic inflammation (TNF-α, IL-1β, IL-6); (iii) sympathetic activation (noradrenaline); (iv) central and peripheral serotonergic signaling (TPH1, TPH2, 5-HTT); and (v) endocrine inputs (including prolactin). Lifestyle modifiers (physical inactivity, alcohol, smoking, and suboptimal diet) are shown as parallel inputs that commonly co-occur with depression. The central part of the panel highlights the principal downstream targets within the osteoblast–osteoclast–osteocyte triad, including the RANK/RANKL/OPG axis, β-adrenergic signaling, and the Wnt/sclerostin pathway. (B) Schematic summary of potential antidepressant-related cellular targets in bone, including osteoclast activity, osteoblast activity, and altered bone-cell signaling, with downstream consequences for BMD and fracture risk. Abbreviations: 5-HTT, serotonin transporter; BMD, bone mineral density; BMI, body mass index; HPA, hypothalamic–pituitary–adrenal; IL, interleukin; OPG, osteoprotegerin; RANK, receptor activator of NF-κB; RANKL, RANK ligand; TNF-α, tumor necrosis factor-alpha; TPH1/TPH2, tryptophan hydroxylase isoforms 1 and 2. Arrows indicate direction of biological influence between drawn entities and do not imply exclusivity, magnitude, or proven causality. ↑ and ↓ indicate the typical reported direction of change in the underlying literature.

Effects of ketamine, its enantiomers, and metabolites on bone health

Beyond classical antidepressants, emerging evidence suggests that ketamine may modulate bone metabolism by altering inflammatory bone markers in patients with MDD. In a clinical biomarker study, individuals with MDD who exhibited low baseline bone-marker levels showed a significant post-infusion increase in the OPG/RANKL ratio and plasma osteopontin (OPN), alongside a decrease in RANKL, following a single intravenous ketamine infusion⁶⁷; in contrast, healthy controls receiving the same infusion displayed no significant changes.⁶⁷ This contrast provides a partial separation between drug and disease, suggesting that ketamine’s acute biomarker effects may depend on the underlying depressive state rather than reflecting a generic pharmacological action on bone. These findings, particularly the reduction in RANKL, are consistent with a transient shift toward a less resorptive bone-turnover profile following acute ketamine exposure; however, given the small single-cohort design and biomarker-only endpoint, this remains a limited/emerging observation requiring independent replication before clinical inferences are drawn. Sustained pharmacological inhibition of RANKL signaling with denosumab increases BMD over months to years,^102,104 providing biological plausibility that RANKL modulation can influence skeletal outcomes; the transient biomarker effect observed after a single ketamine infusion, however, does not justify direct extrapolation to clinical bone protection.

A more recent open-label clinical biomarker study extends this question to repeated ketamine exposure. In 102 patients with unipolar or bipolar depression receiving 6 intravenous ketamine infusions over 12 days, Lan et al. measured 8 plasma bone-related markers at baseline, 24 hours after the sixth infusion and again 2 weeks later. Repeated ketamine infusions were associated with increases in leptin and OPG and decreases in osteocalcin, sclerostin, OPN, PTH, and FGF23 at post-infusion time points, with sex-specific differences in some marker trajectories.¹⁵⁴ These findings support the view that ketamine exposure can modulate bone-related biomarkers in patients with depression, but it is important to emphasize that these biomarker-level observations are from an open-label study that does not establish effects on BMD, bone microarchitecture, or fracture risk.

Not all inflammatory mediators respond to ketamine in the same way. One clinical study found that levels of TNF-α, interferon-γ (IFN-γ), IL-2, IL-5, and IL-10 remained unaffected after ketamine administration, whereas IL-6 increased approximately 4 hours post-infusion.¹⁵⁵ These cytokine changes were not correlated with antidepressant response, suggesting that they do not constitute the primary mechanism underlying ketamine’s rapid mood effects.¹⁵⁵ Preclinical studies further suggest enantiomer-specific effects. In a chronic social defeat stress (CSDS) mouse model, arketamine, but not esketamine, reversed stress-induced reduction in the OPG/RANKL ratio, with this ratio positively correlating with sucrose preference, an antidepressant-relevant behavioral measure.¹⁵⁶

Consistent with this preclinical biomarker signal, arketamine has been reported to attenuate BMD deficits in an ovariectomized (OVX) mouse model of estrogen-deficiency-induced bone loss,¹⁵⁷ and repeated intermittent administration of arketamine, but not esketamine, ameliorated reductions in cortical and total BMD over 6 weeks in OVX mice.¹⁵⁸ Because the OVX model induces bone loss through estrogen deficiency rather than a depression-like phenotype, these findings provide some separation between arketamine’s action on bone and depression-mediated indirect pathways, although they remain confined to female rodents, short follow-up windows, and preclinical endpoints, and have not been replicated in humans. Research examining skeletal effects of arketamine and esketamine remains limited, and studies on (2R,6R)-HNK, a ketamine metabolite under clinical development for depression treatment,¹⁵⁹ are even scarcer. One investigation using the CSDS model reported that (2R,6R)-HNK failed to reverse bone loss or normalize the OPG/RANKL ratio, suggesting that it may retain antidepressant-relevant efficacy without reproducing the bone-related effects observed with arketamine.¹⁶⁰ Direct in vitro studies of arketamine, esketamine, and (2R,6R)-HNK at clinically relevant concentrations remain lacking, although racemic ketamine has been shown in 1 study to inhibit osteoclast differentiation and function through suppression of nuclear factor of activated T cells 1 (NFATc1) and (Fos proto-oncogene, AP-1 transcription factor subunit) c-Fos pathways.¹⁶¹

Additional preclinical evidence also cautions against assuming uniformly favorable skeletal effects of ketamine. In a rat bone-regeneration context, Horváthy et al. reported that repeated ketamine–xylazine anesthetic exposure was associated with impaired calvarial bone healing, while in vitro ketamine exposure reduced viability of spongious and cortical bone explants in a concentration-dependent manner, with significant effects at 2 mM.¹⁶² These findings differ from antidepressant-dose ketamine studies because they involve anesthetic administration, xylazine co-exposure, bone-regeneration endpoints, and high in vitro concentrations; nevertheless, they reinforce that ketamine-related skeletal effects should be interpreted as model-, dose-, and exposure-dependent rather than uniformly bone-protective.

Taken together, the current evidence base on ketamine, its enantiomers, and its metabolites in relation to skeletal health is hypothesis-generating rather than confirmatory. It comprises 2 human biomarker studies, including a single-infusion study in MDD and an open-label repeated-infusion study in unipolar and bipolar depression,^67,154 a small number of rodent studies suggesting that arketamine—but not esketamine or (2R,6R)-HNK—attenuates stress- and OVX-induced bone loss in female mice over short follow-up windows,^{155–158,160} and limited in vitro/preclinical evidence indicating that racemic ketamine can alter osteoclast biology and, under some anesthetic or high-concentration conditions, impair bone-cell viability, or bone regeneration.^161,162 To our knowledge, no clinical longitudinal BMD studies and no fracture-outcome studies of ketamine, esketamine, or arketamine have been published. Claims of clinical bone protection in humans are therefore not supportable at present, and the biomarker, rodent, and in vitro findings summarized above should be regarded as motivating signals for adequately powered prospective studies with longitudinal bone-imaging, falls, and fracture endpoints.

Clinical monitoring and management implications

Fracture risk in patients receiving antidepressants is likely mediated through at least 2 partially independent pathways: effects on skeletal metabolism and an increased propensity for falls. Antidepressant-related contributors to falls include orthostatic hypotension, particularly with TCAs and certain SNRIs, psychomotor slowing, sedation, impaired attention, sleep disturbances, movement disorders, and SSRI-associated hyponatremia, especially in older adults, all of which may adversely affect postural stability and cognitive function.¹⁶³ Depression itself is independently associated with increased falls risk through psychomotor retardation, reduced physical activity, sarcopenia, frailty, impaired executive function, and multimorbidity.¹⁶⁴ Consequently, fracture signals observed in epidemiological studies of antidepressant use are likely to reflect a combination of skeletal fragility and falls-related mechanisms. Studies focusing exclusively on BMD may therefore underestimate clinically meaningful fracture risk, whereas fracture-endpoint studies cannot readily distinguish between fractures attributable to reduced bone strength and those resulting from increased falls propensity. This distinction highlights the importance of interpreting observational antidepressant–fracture associations cautiously and considering frailty, functional status, and falls risk as major confounding variables rather than secondary considerations.

Currently, no major osteoporosis guideline recommends routine BMD assessment solely on the basis of antidepressant exposure, and the evidence reviewed herein does not support incorporating antidepressant use as an independent osteoporosis screening criterion. Contemporary fracture-risk assessment frameworks and screening recommendations remain centred on established risk factors including age, sex, menopausal status, low BMI, prior fragility fracture, parental hip fracture, smoking, glucocorticoid exposure, alcohol intake, rheumatoid arthritis, and other recognized secondary causes of osteoporosis.^165–167 Importantly, neither antidepressant exposure nor falls history currently constitute formal Fracture Risk Assessment Tool (FRAX) variables despite accumulating evidence linking both factors to fracture outcomes.

A pragmatic interpretation of the available evidence is therefore to consider long-term antidepressant exposure, particularly prolonged SSRI treatment, as a contextual fracture-risk modifier rather than an independent indication for dual-energy X-ray absorptiometry (DXA). Rather than serving as a standalone criterion for BMD assessment, antidepressant exposure may reasonably contribute to clinical decision-making when the overall fracture-risk profile is already borderline or elevated. Such situations may include (1) postmenopausal women receiving long-term SSRI therapy who also present with low BMI, smoking history, or previous fragility fracture; (2) older adults receiving systemic glucocorticoids, which represent a recognized FRAX risk factor, or other medications with established or suspected skeletal effects, including proton-pump inhibitors¹⁶⁸ and selected anticonvulsant or antiepileptic agents¹⁶⁹; and (3) individuals with a family history of osteoporosis, recurrent falls, frailty, or additional clinical indicators suggestive of elevated fracture risk. In such circumstances, DXA/BMD assessment may provide clinically useful information within a broader risk-stratification framework. Decisions regarding DXA referral should therefore be individualized and guided by comprehensive fracture-risk assessment rather than antidepressant exposure alone.

For patients with established osteoporosis or high falls risk, skeletal status and functional capacity may reasonably inform antidepressant selection and monitoring strategies. However, current evidence remains insufficient to support preferential prescribing of 1 antidepressant class over another exclusively on skeletal grounds. Psychiatric indication, treatment efficacy, tolerability, and patient-specific clinical needs should remain the primary determinants of antidepressant choice.

Exercise, physiotherapy-informed rehabilitation, and mechanical loading represent clinically actionable adjunctive strategies because their beneficial effects on skeletal health, falls prevention, physical function, and depressive symptoms are supported by substantial evidence from osteoporosis, rehabilitation, and psychiatric literature.^170–174 Current position statements recommend multicomponent exercise programs that incorporate weight-bearing impact activities when safe, progressive resistance training, balance training, and functional mobility exercises tailored to age, baseline physical capacity, fracture history, comorbidities, and falls risk.^170,171 Simultaneously, meta-analytic evidence demonstrates that structured exercise interventions significantly reduce depressive symptom severity, with combined exercise and standard treatment approaches producing greater antidepressant effects than standard treatment alone.^172,173

Given that individuals with depression often exhibit reduced physical activity levels and may experience motivational, cognitive, and fatigue-related barriers to exercise participation, rehabilitation programmes should adopt a gradual and individually tailored progression model supported by behavioural adherence strategies. Practical implementation may utilize the frequency, intensity, time, and type (FITT) framework,¹⁷⁵ integrating weight-bearing, resistance, and balance-training components. Physiotherapy-led assessment of gait, lower-limb strength, postural control, functional mobility, and falls history may be particularly valuable in older adults and other high-risk populations.^176,177

Nutritional assessment should likewise form part of an integrated skeletal-health strategy. Particular attention should be paid to adequate energy availability, protein intake, fruit and vegetable consumption, and vitamin D/calcium sufficiency.¹⁷⁸ This consideration is especially relevant given that chronic stress and depressive disorders can substantially alter appetite regulation, dietary quality, and body weight, each of which may influence bone remodeling and skeletal integrity.^125,179 Higher fruit and vegetable consumption has consistently been associated with greater BMD and lower osteoporosis risk, potentially through the provision of micronutrients, antioxidants, and bioactive compounds that support osteoblast function and suppress excessive osteoclastic activity.¹⁸⁰ Conversely, stress-related reductions in appetite and body weight may adversely affect skeletal health through both nutritional deficits and reduced mechanical loading.¹²⁵

The mechanistic rationale for movement-based interventions extends beyond skeletal loading alone and may intersect with several biological pathways implicated in the depression–bone relationship. As discussed throughout this review, sympathetic nervous system activation, β-adrenergic signaling, HPA axis dysregulation, and chronic low-grade inflammation represent plausible mechanistic links between depression and altered bone remodeling.^20,125,133 Structured exercise has been shown to influence autonomic regulation, attenuate HPA-axis hyperactivity, and modulate systemic inflammatory signaling across diverse clinical populations, including individuals with depressive disorders,^181,182 while specific physiotherapy techniques can also acutely modify sympathetic indices in experimental settings.^183,184 Collectively, these observations provide a biologically plausible framework through which exercise may influence both mental and skeletal health. Nevertheless, direct evidence linking exercise-induced changes in autonomic, neuroendocrine, or inflammatory pathways to improvements in bone outcomes among patients with depression remains limited. Future longitudinal studies should therefore integrate fracture outcomes, BMD, and bone microarchitecture measurements with autonomic, endocrine, and inflammatory biomarkers in patients with depression receiving combined pharmacological and exercise-based interventions.

Importantly, although emerging preclinical and translational evidence suggests that ketamine-related compounds may influence bone-remodeling pathways, no longitudinal clinical studies have yet evaluated the effects of ketamine, esketamine, arketamine, or (2R,6R)-HNK on BMD, bone microarchitecture, skeletal strength, or fracture incidence. Consequently, no ketamine-specific skeletal monitoring recommendations can currently be made beyond standard osteoporosis risk assessment procedures. Addressing this knowledge gap should represent a priority for future research as rapid-acting antidepressants become increasingly integrated into routine psychiatric practice. The clinical considerations summarized in Table 3 should be interpreted as an evidence-informed synthesis of the literature reviewed in this manuscript rather than as formal clinical guidelines. Until prospective longitudinal studies provide stronger evidence regarding the skeletal consequences of antidepressant exposure, clinicians should adopt an individualized risk-assessment approach that integrates psychiatric status, treatment duration, fracture-risk profile, falls propensity, functional capacity, and overall health status when considering skeletal monitoring and preventive interventions.

Table 3.

Practical clinical monitoring considerations for skeletal health in patients receiving long-term antidepressant treatment

Monitoring/management area	When to consider	Practical consideration
Fracture-risk assessment	Older age, postmenopausal status, low BMI, prior fragility fracture, family history, smoking, alcohol use, glucocorticoid exposure	Use established fracture-risk assessment; antidepressant exposure should be considered a contextual risk modifier, not a standalone indication for DXA
BMD/DXA assessment	Long-term SSRI therapy plus additional osteoporosis or fracture-risk indicators; borderline or elevated overall fracture-risk profile	Antidepressant exposure should not be treated as a standalone DXA indication but may support DXA/BMD assessment when the overall clinical fracture-risk profile is already borderline or elevated
Falls and frailty evaluation	Older adults, prior falls, orthostatic symptoms, sedation, psychomotor slowing, sarcopenia, hyponatremia risk	Assess falls history, gait, balance, orthostatic symptoms, functional mobility, and medication burden
Antidepressant treatment selection	Established osteoporosis or high falls risk	Bone/falls profile may inform treatment choice but should not override psychiatric indication, efficacy, or tolerability
Physical activity and rehabilitation	Low activity, weakness, impaired balance, fear of falling, or functional limitation	Consider weight-bearing/resistance exercise, balance training, and physiotherapy-informed rehabilitation where appropriate
Nutrition and vitamin D/calcium status	Poor diet, weight loss, low protein intake, vitamin D/calcium insufficiency	Address energy intake, protein, fruit/vegetable intake, and vitamin D/calcium sufficiency as part of overall risk management
Rapid-acting antidepressant exposure	Ketamine, esketamine, or arketamine treatment	No ketamine-specific skeletal monitoring recommendation can currently be made; longitudinal BMD and fracture data are lacking

These considerations are intended as a practical synthesis of monitoring issues and should not be interpreted as formal clinical guidelines or as standalone screening recommendations. Antidepressant exposure, particularly long-term SSRI therapy, may be considered as a contextual risk modifier within established fracture-risk assessment, but current evidence does not support routine DXA screening based on antidepressant exposure alone.

Abbreviations: BMD, bone mineral density; BMI, body mass index; DXA, dual-energy X-ray absorptiometry; SSRI, selective serotonin reuptake inhibitor.

Limitations and future directions

There is considerable progress in establishing depression as a factor related to poorer bone health; however, methodological issues still limit the strength of this inference. Methods for assessing depression considerably vary, with some studies using structured diagnostic interviews and others relying on self-report symptom scales, resulting in inconsistent classification across different cohorts and studies. Similarly, bone status is most often evaluated by DXA, which is extensively used but has limited sensitivity for detecting trabecular bone changes, and thus, it may underestimate the actual extent of bone loss.¹⁸⁵ Future studies should include additional endpoints such as fracture incidence, bone microarchitecture assessed by high-resolution peripheral quantitative computed tomography, and other indices of bone quality.

Other sources of variability include small sample sizes, limited consideration of sex, menopausal status and baseline fracture-risk factors, and underrepresentation of diverse ethnic and geographic populations. To further advance the field, large prospective longitudinal cohorts with harmonized measures of depression and antidepressant exposure, as well as skeletal outcomes need to be conducted, along with randomized controlled trials comprising placebo arms and head-to-head comparisons across different antidepressant classes wherever it is ethically and clinically appropriate. Addressing these limitations will strengthen the evidence base and help clarify how depression and its medical treatment interact to influence long-term skeletal health.

Currently available human data are largely observational, which limits causal inference. Confounding is consistently incomplete. Indeed, relatively few primary studies simultaneously adjust for depression severity and chronicity, frailty and falls propensity, smoking, alcohol consumption, BMI and overall nutritional status, vitamin D and calcium intake, current or prior glucocorticoid exposure, menopausal status and gonadal hormone exposure, and the broader burden of medical comorbidity. Confounding by indication is a critical threat to validity in analyses of antidepressant exposure because the underlying depression that prompts treatment is itself a determinant of bone health; analyses that report associations between antidepressant use and BMD or fracture risk without depression-severity adjustment or appropriate comparator selection cannot reliably separate drug effects from disease effects. Exposure phenotyping is similarly incomplete, with class, dose, duration, adherence, and switching insufficiently captured, and class-specific effects of SSRIs, SNRIs, and TCAs not clearly separated from the influence of MDD itself. Outcomes rely heavily on DXA, which can miss trabecular loss and microarchitectural change; future work should add high-resolution peripheral quantitative computed tomography, vertebral fracture assessment, finite-element strength estimates, and standardized bone-turnover markers with attention to preanalytical control. Together, these limitations explain a substantial proportion of the inter-study heterogeneity highlighted in Table 2.

An important limitation in the available evidence reviewed here is the lack of sex-, gender-, and age-inclusive data across the depression–bone literature. The majority of clinical studies have been conducted in postmenopausal women, in part because this group carries the highest baseline osteoporosis risk. Consequently, premenopausal women, men across the lifespan, and sex-diverse populations remain underrepresented, and few studies disaggregate or report outcomes by sex or gender identity. Age coverage is similarly uneven: Data on children and adolescents, in whom antidepressant exposure may coincide with peak bone accrual, are sparse, while the oldest-old, who carry the highest fracture and falls burden, are also underrepresented in trials. As a result, the present evidence may obscure clinically important sex- and age-specific effects of depression and antidepressant exposure on skeletal health. Future studies should adopt sex- and gender-disaggregated reporting from the outset, ensure adequate recruitment across the lifespan, and explicitly include men and gender-diverse participants alongside the more commonly studied postmenopausal cohorts.

Mechanistic translation is also limited. Direct in vitro testing of osteoblast and osteoclast responses to ketamine enantiomers and (2R,6R)-HNK at clinically relevant concentrations is scarce, and in vivo dosing should be aligned to human pharmacokinetics. Biomarker panels such as OPG/RANKL/OPN and cytokines need validation as predictive rather than merely associative signals of skeletal change during antidepressant treatment, ideally integrated with multi-omics, including microRNA profiles, and imaging endpoints. A critical gap is that no published study has measured longitudinal BMD, bone microarchitecture, or fracture incidence in patients receiving ketamine, esketamine, or arketamine. Existing human studies remain limited to bone-related biomarker outcomes, and preclinical findings are model-, dose-, and exposure-dependent; until longitudinal skeletal endpoint data become available, the findings reviewed in the “Effects of Ketamine, Its Enantiomers, and Metabolites on Bone  Health” section should be treated as hypothesis-generating rather than as evidence of clinical bone protection.

Priority next steps include randomized or pragmatic trials with skeletal endpoints for long-term SSRI or SNRI exposure and, where clinically justified, head-to-head mechanistic comparisons of arketamine and esketamine with longitudinal bone-imaging outcomes. Real-world registries for ketamine and esketamine should incorporate bone surveillance, fracture outcomes, falls, and frailty measures. Prospective studies should apply causal-inference methods, prespecify stratification by BMI, inflammatory burden, vitamin D and calcium status, and key co-medications, and systematically capture lifestyle factors, in addition to the sex-, gender-, and age-related considerations highlighted above. Standardization and transparency, including protocol preregistration, harmonized case definitions using clinician interview versus self-report, FAIR data practices, and reporting of neutral results, will be essential to strengthen the evidence base and guide risk-mitigation strategies.

Conclusions

The evidence reviewed in this article identifies several consistent associations and plausible converging mechanisms linking depression, its pharmacological treatment, and skeletal health. It also makes clear that important uncertainties remain. MDD is consistently associated in observational human studies and complementary preclinical models with reduced BMD and increased risk of osteoporosis and fracture.^114–116 Plausible mechanisms include HPA-axis hyperactivation with sustained cortisol elevation, suppression of osteoblast differentiation and survival, shifts in the RANKL/OPG balance toward osteoclastogenesis, stress-induced increases in pro-inflammatory cytokines, sympathetic overactivity, receptor- and context-dependent effects of peripheral 5-HT signaling on bone-cell function, Wnt/sclerostin pathway regulation, and prolactin-mediated modulation of osteoblast cytokine output.^{111,126,127,130,131,133,137,138,150,151}

Classical antidepressants, particularly SSRIs, have been associated with adverse bone outcomes in observational human studies and with direct effects on osteoblast and osteoclast biology in vitro.^25,148,149 These associations remain difficult to interpret causally because depression severity, frailty, falls risk, lifestyle factors, and comorbidity are inconsistently controlled.

For ketamine, the available human evidence remains limited to biomarker-level studies, including a single-infusion study in MDD and an open-label repeated-infusion study in unipolar and bipolar depression, both reporting changes in bone-related markers rather than longitudinal skeletal endpoints.^67,154 Preclinical rodent evidence indicates that arketamine, but neither esketamine nor (2R,6R)-HNK, attenuates stress- and ovariectomy-induced bone loss over short follow-up periods.^{156–158,160} However, other preclinical data indicate that racemic ketamine can alter osteoclast biology and may impair bone-cell viability or bone regeneration under anesthetic or high-concentration experimental conditions.^161,162 In the absence of clinical longitudinal BMD, bone microarchitecture, or fracture data, these findings remain hypothesis-generating and do not support an inference of clinical bone protection in humans.

Given the current state of the evidence, bone health should be considered as part of the long-term management of patients with MDD, particularly in those with additional osteoporosis risk factors such as postmenopausal status, older age, chronic glucocorticoid exposure, low BMI, smoking, or excess alcohol use. Modifiable lifestyle factors, including physical activity, nutrition, vitamin D and calcium status, smoking, and alcohol intake, should be addressed alongside pharmacotherapy.^101,102 Periodic BMD assessment may be reasonable in higher-risk patients receiving long-term SSRI therapy, with osteoporosis treatment initiated according to established fracture-risk guidance; current evidence does not, however, support avoiding clinically indicated antidepressant treatment or preferentially prescribing 1 antidepressant class over another on bone-health grounds alone. Finally, no clinical recommendation regarding ketamine, esketamine, or arketamine for skeletal indications is supportable at present. The preclinical and biomarker signals reviewed here are best viewed as motivating questions for adequately powered prospective human studies with longitudinal bone-imaging, falls, and fracture endpoints.

Author contributions

Polymnia Louka (Conceptualization [equal], Writing—original draft [lead], Writing—review & editing [equal], Funding acquisition), Andrew Pitsillides (Writing—original draft [supporting], Writing—review & editing [supporting]), Morfeas Koumas (Visualization [supporting, Figure 1 prepared under the guidance of P.L.], Writing—original draft [supporting], Writing—review & editing [supporting]), Charalambos Papacharalambous (Conceptualization [supporting, Clinical Monitoring and Management Implications section and Table 3], Visualization [supporting, graphical abstract generation], Writing—review & editing [equal], critical clinical interpretation of evidence throughout the revised manuscript), Antonia Sophocleous (Writing—original draft [supporting], Writing—review & editing [supporting]), and Panos Zanos (Conceptualization [equal], Supervision [lead], Visualization [supporting], Writing—review & editing [equal], Funding acquisition).

Conflicts of interest

P.Z. is listed as co-inventor in patents and patent applications related to the pharmacology and use of (2R,6R)-hydroxynorketamine in the treatment of depression, anxiety, anhedonia, suicidal ideation and post-traumatic stress disorders. All other authors report no conflict of interest.

Funding

This work was supported by the Research and Innovation Foundation of Cyprus [grant#: EXCELLENCE/0421/0543] to P.Z. and the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie Grant Agreement No. 101034403 to P.L.

Data availability

No new data were generated in support of this work.