The present study investigated the BP effects of trazodone in a sample of older hypertensive outpatients. The study results showed that trazodone use was associated with a greater systolic and diastolic BP reduction immediately after standing and a possible increase in risk of syncope and falls, independently of age, recent fall history and disability.

Trazodone is an antidepressant medication with a dual mechanism of action involving inhibition of the serotonin transporter (SERT) and serotonin type 2 receptor (5-HT2A and 5-HT2C receptors). Moreover, it exerts antagonistic properties against histamine H1 receptor and α1- and α2-adrenergic receptors [4, 23]. Previous studies report OH as one of the adverse effects of trazodone therapy, especially common in older adults and in patients with pre-existing cardiac disease [5, 6]. However, to the best of our knowledge, this is the first study to explore the effects of trazodone on different BP measurements, including a detailed analysis of the orthostatic BP response.

The hypotensive effect of trazodone is primarily attributed to the inhibition of α1-adrenergic receptors [23]. It is usually transient and dose-related, being less pronounced when low doses or prolonged-release formulations are used [4]. The inhibition of α-adrenergic receptors provides a likely explanation for the greater BP drop reported among trazodone users in our study sample. A similar effect has been described for benzodiazepines, although different underlying mechanisms have been hypothesised [24]. Trazodone users also showed lower office diastolic BP consistently, with a possible vasodilating effect deriving from α-adrenergic blockade. We cannot exclude that this BP profile may have been influenced also by other predisposing factors for OH which commonly coexist in older adults, such as advanced age, dementia, polypharmacy and physical deconditioning. In particular, the association of trazodone with orthostatic BP drop may be partly explained by the higher prevalence of dementia among trazodone users, resulting in orthostatic hypotension owing to autonomic dysfunction. Consistently, prevalence of OH was relevant in our study (43.8%), as it might be expected in a sample of frail older patients.

The present study also identified a significant association between trazodone use and the risk of syncope and falls, which was at least partly independent of other predisposing factors such as old age, fall history and disability. While no study has investigated the association between trazodone and syncope, our findings agree with previous data describing an increased risk of falls in community-dwelling older adults and long-term care residents receiving trazodone [25,26,27,28]. In particular, some evidence suggests that trazodone may carry a similar fall risk compared with benzodiazepines and zopiclone in frailer individuals such as nursing home residents [28, 29]. Of note, treatment with benzodiazepine and antipsychotics was not associated with fall risk in the present sample. Fall risk associated with trazodone is typically attributed to daytime drowsiness and dizziness deriving from its sedative effects. The impact of trazodone on orthostatic BP observed in the present study suggests that also BP fluctuations may play a role; however, we reported no association between BP drop and the study outcome. Moreover, the association between trazodone and fall risk appears weakened when dementia was considered as a confounder, suggesting that trazodone use is not a risk factor per se and that the association results from the cumulative effect of several risk factors. In fact, fall susceptibility in older individuals derives from the synergic effects of multiple predisposing factors, such as medications, cognitive and physical impairment, BP and sensory deficits [30, 31]. The association between trazodone use and fall risk should thus be interpreted in the context of this multifactorial phenomenon, taking into consideration that multiple different conditions may modulate each individual’s risk of falling. On the basis of our findings, we may hypothesise that trazodone may interfere with the BP response to standing and potentially contribute to increase the risk of syncope and falls in older adults with greater cumulative fall risk, especially older subjects with dementia, in whom trazodone use is more common. Therefore, OH and fall risk should be carefully investigated before trazodone initiation, and modifiable risk factors should be addressed to minimise the risk of fall-related complications. Moreover, the use of prolonged-release formulations and/or fractionate doses may be preferred, given the lower BP effects [4].

Some limitations of the present study should be considered. First, detailed information on trazodone dosage, formulation (e.g. prolonged release formulations) and treatment duration (new/chronic use) was not available in our dataset. Second, details on the circumstances and characteristics of syncope and falls were not recorded, and we were unable to draw any conclusion on the aetiology of events or on their correlation with BP values and trazodone use. Third, our results may not apply to older adults from other clinical settings, e.g. community-dwelling subjects of younger ages or individuals with lower frailty and disability level, that may show a lower risk of hypotension and falls. Finally, the small sample size and the low number of trazodone users (N = 12, 9.8% of the whole sample) may have influenced the analysis of BP values and did not allow for a subgroup analysis investigating trazodone effects in subjects with a different predisposition to hypotension and falls. Moreover, we cannot exclude that the low number of trazodone users might indicate a selection bias.