Balancing bone health and mental health: a new clinical dilemma?

The postmenopausal woman who has been treated for depression with paroxetine over several years presents a new clinical dilemma. Should depression or treatment with antidepressant medications be cause for concern regarding the health of her bones? Does her depressive condition and/or its treatment contribute to her existing risk factors for fracture and bone loss?

Interest in the effects of serotonin on bone health has been spurred by recent reports of lower bone mineral density (BMD), increased bone loss and increased fracture rates among selective serotonin reuptake inhibitor (SSRI) users. Three separate cohort studies in both men and women have shown evidence of a clinical effect of SSRIs on bone outcomes. Men who reported SSRI use had a lower BMD compared with those using other antidepressants and those using no antidepressant medication in a multicenter cohort study of older men, the Osteoporotic Fractures in Men (MrOS) study [1]. A complementary finding was demonstrated in women from the Study of Osteoporotic Fractures (SOF). Women using SSRIs in that study demonstrated increased rates of bone loss over 4 years compared with nonusers [2]. A third cohort from the Canadian Multicenter Osteoporosis Study (CaMos) demonstrated low BMD, as well as increased fracture rates among men and women who reported SSRI use at a single time point [3].

Prior to studies that compared bone outcomes among SSRI users and nonusers, reports of fractures and lower BMD among depressed patients raised questions about the relationship between depression and bone health [4–9]. Participants in the studies of depression may or may not have been taking SSRIs. Some studies were published before SSRI use was widespread and others failed to report medication use among the population. There are theoretical reasons that depression and bone health might be related, but the direction of such an association has not been clear – that is, whether depression could lead to osteoporosis or osteoporosis could lead to depression. Potential mechanisms may be biological, physiologic or mediated at the cellular level by the serotonin transporter.

From a biologic standpoint, it is conceivable that people who receive a diagnosis of osteoporosis or sustain a fracture, would become depressed as a result of pain related to a fracture, by new limitations in function or simply by having a diagnosis of a chronic disease. Depression is associated with other chronic medical conditions [10,11], each of which might affect bone health individually and in aggregate. Physical activity is an important contributor to bone health, and depression may lead to lack of motivation, loss of interest in one’ usual activities and a reduction in physical activity. Depression may lead to behavior changes that result in poor nutrition, reduced sunlight exposure and low vitamin D levels, which can contribute to reduced bone health.

Pathophysiologic changes related to depression may contribute to bone loss and fracture risk. Increased cortisol levels among depressed individuals have the potential to accelerate bone loss [12] through mechanisms akin to those seen in patients with Cushing’s syndrome or those using long-term systemic glucocorticoid medications. Likewise, elevations in cytokines could also contribute to an increase in fracture rate and bone loss [13].

The serotonin transporter provides a potential mechanism to explain the association of SSRI use and bone loss/fracture risk at the cellular level. SSRIs inhibit the serotonin transporter centrally and peripherally, in what is thought to be their mechanism of action for treating depression. Serotonin transporters are expressed in bone cells (osteoblasts, osteoclasts and osteocytes) and disruption of the serotonin transporter gene results in lower bone mass, altered bone architecture and reduced bone strength in mice [14–17]. Thus, the clinical findings that SSRI use is associated with lower bone density, bone loss and fractures, are supported by convincing in vivo and in vitro evidence that the serotonin transporter may have direct effects on bone physiology.

Whether depression or the medications used to treat depression (or both) impact bone health is the central question driving recent studies of SSRIs. Unfortunately, the studies published thus far, are limited by the potential for confounding by indication. Confounding by indication can occur in epidemiologic studies when both a medical condition and a medication used in its treatment have potential to be associated with the outcome in question [18,19]. In the case of serotonin and bone, theoretical mechanisms exist for how depression and antidepressant medications (particularly SSRIs) might contribute to low BMD, bone loss and fractures. All of the recent studies that linked SSRIs and osteoporosis attempted to account for confounding by indication, by adjusting for depressive symptoms, but none have included rigorous diagnostic interviews for depression [1–3].

Clinicians faced with trying to incorporate these new findings into their clinical practice may struggle with deciding how to balance the risks and benefits of SSRI treatment for depression. Depression remains an important cause of disability worldwide, a leading cause of medical morbidity and an important contributor to poor outcomes in managing other chronic diseases [20]. Since their introduction, SSRIs have been considered the safest of the antidepressant medication choices [21]. When compared with older classes of antidepressant medication, they carry a lower risk for drug–drug interactions, have a lower fall risk and are not fatal when taken in overdose. For people with major depressive disorder, SSRIs are likely to continue to be first-line treatment based on safety and efficacy.

The number of people taking SSRIs for disorders other than major depressive disorder is unknown. SSRIs are US FDA approved for major depressive disorder, anxiety, post-traumatic stress disorder and, recently, premenstrual dysphoric disorder. In practice, they may be used for other, less-severe, mood disturbances. This proliferation of SSRI use for varied indications will increase the difficulty in calculating risk–benefit ratios in the context of maintaining bone health. The risk–benefit ratio for a person with major depressive disorder who is treated with an SSRI and doing well is likely to favor continuing treatment with SSRIs. However, the same calculation for a person with a less-severe disorder, or an unapproved or unstudied indication, makes a less compelling case for continuation of SSRI treatment. Decisions to stop treatment should not be made without the input of the prescribing physician, after careful consideration and with monitoring.

When the risk–benefit calculation favors continuation of SSRI treatment, earlier testing with bone density measurements and assessment of vitamin D stores may be appropriate. Heightened awareness of the potential need to screen for the usual secondary causes of osteoporosis that may exist in a depressed patient is reasonable. Attention to preventive measures, such as calcium supplementation and exercise, are always important but deserve particular attention among SSRI users. Since the in vivo studies in young mice indicate that the mechanism of reduced bone accrual in SSRI-treated animals is a reduction in bone-formation rates without concomitant increased resorptive activity, it is unclear whether an antiresorptive drug, such as a bisphosphonate, or an anabolic agent (e.g., teriparitide) would be the most appropriate choice for treating bone loss associated with SSRI use.

Thus, concern about negative effects of SSRIs on bone will challenge clinicians to reconsider whether treatment is warranted for some individuals, and seek to minimize risk in those with clear indications for SSRI treatment. As we await further evidence to direct these decisions, clinicians may struggle with this dilemma. All medications have risks and benefits and it is the clinician’s art to balance these for each individual patient.

Several opportunities exist for further investigation of the relationship between SSRIs, depression and bone outcomes. Future randomized controlled trials of antidepressant medications, especially those that target the serotonin transporter, should include bone outcomes as part of their monitoring for adverse effects. These studies should make an attempt to definitively assess and control depression in order to account for confounding by indication. Demonstration of a dose–response effect by SSRIs would further support a cause–effect relationship between SSRIs and osteoporosis. Assessment of bone outcomes in previous randomized controlled trials of SSRIs (via fracture follow-up or retrospective biomarker of bone turnover analysis) would be helpful, but would require collaboration from pharmaceutical companies who may not wish to explore potential toxicities in approved drugs.

Just as studies are necessary to further our clinical understanding of the relationship between SSRIs, depression and bone health, so are studies to elucidate the exact nature of cellular mechanisms for the effect of serotonin on bone. Understanding and describing the serotonin receptor and transporter signaling pathways in osteoblasts, osteoclasts and osteocytes will be important. Not only will this help with understanding the pathophysiology of osteoporosis, but it may help target future treatments both for osteoporosis and depression. Development of medications that preferentially target serotonin transporters that directly affect depression, but have little activity on bone would be desirable.

In conclusion, recent reports suggest an effect of SSRIs on bone that is not neglibile. Decisions regarding treatment for depression, as with all medical conditions, must weigh the risks and benefits. There is an urgent need for carefully designed clinical and basic science studies to improve the understanding of the relationship between the nervous system and bone in general and, more specifically, the impact of SSRIs on bone.

Acknowledgements

Thank you to Christina Nicolaidis and M Michael Bliziotes for their suggestions and review of this manuscript.

Financial & competing interests disclosure

EM Haney is supported by a grant from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (K23 AR 051926). The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.