Drug-related problems (DRPs) are major health problems.1,2 Oral anticoagulants (OACs) are one of the therapeutic groups most frequently involved in these DRPs, being gastrointestinal bleeding the main cause of ED consultation.3 Gastrointestinal bleeding occurs in 8% to 15%4,5 of patients treated with vitamin K antagonists (VKAs), with earlier studies not showing significant differences when compared to direct-acting oral anticoagulants (DOACs). However, more recent evidence indicates that the incidence of gastrointestinal bleeding varies among individual DOACs, with apixaban being associated with a lower risk and rivaroxaban with a higher risk of gastrointestinal hemorrhage.6

Antidepressants, in particular selective serotonin reuptake inhibitors (SSRIs), are used for a variety of indications such as depression, anxiety or pain disorders. Over the last decades, there has been an increase in the rates of prescription of antidepressant medications across all ages, with the largest rise reported in older adults, being prescribed in about 30% of elderly patients.7 Recently, more attention has been given to SSRIs surrounding the risk of bleeding.8 Bleeding associated with SSRI use has been associated with the inhibition of the serotonin transporter on platelets, leading to reduced platelet aggregation.9 Based on the mechanism, any medication that involves the inhibition of serotonin reuptake would put patients at an increased risk of abnormal bleeding.10 The increased risk of major bleeding when combining SSRIs with OACs may be explained by several mechanisms. During hemostasis, platelets release serotonin to enhance activation and aggregation, promoting coagulation cascade interactions. SSRIs block the serotonin reuptake transporter in the platelet membrane, which decreases the serotonin content of platelets and reduces the efficiency of this process.11 In addition, some SSRIs such as fluoxetine and fluvoxamine inhibit the CYP1A2 and CYP2C9 hepatic isoenzymes of cytochrome P450, which are key determinants in the metabolism of VKA.12

Although previous studies have evaluated the risk of bleeding from the combination of anticoagulants with antidepressants,13,14 the risk in those patients who visit the emergency room with a first episode of bleeding has not yet been established. Moreover, data regarding the differences between patients taking Vitamin K antagonist (VKA) and direct oral anticoagulants (DOACs) and the type of antidepressant are still controversial. Given the increase in the use of anticoagulants among the elderly population,15,16 the identification of risk factors associated with new bleeding episodes is particularly relevant.

Thus, we proposed the present study to evaluate the risk of ED revisit among elderly patients with gastrointestinal bleeding episodes secondary to anticoagulant treatment, including the combination with antidepressants.

A total of 143 patients were included, with a mean age of 82.4 (SD:9.8) years. The baseline characteristics of the patients included are shown in Table 1. Of the total cohort, 51.7% were women and 43.7% were on severe polypharmacy. The most prevalent comorbidities in the subjects were arterial hypertension (82.5%), congestive heart failure (32.2%) and chronic renal failure grade III or higher (32.2%). The proportion of patients with cognitive impairment or Alzheimer's disease was 24.4%.

In terms of treatment, 75.5% of patients were discharged on oral anticoagulation with direct anticoagulants (DOACs), while 24.4% were discharged on vitamin K antagonists (VKA). Antidepressant therapy was prescribed in 26.7% of the patients, with selective serotonin reuptake inhibitors (SSRIs) being the most commonly prescribed (12.6%). Antiplatelet agents were used in 8.3% of cases, and 77.6% of patients received proton pump inhibitors (PPIs). The mean follow-up duration was 90 days (SD: 24.3), and the all-cause mortality during this period was 8.3%.

A total of 33 patients (23.1%) revisited the emergency department due to new episodes of gastrointestinal bleeding within 90 days post-discharge. Fig. 1 shows the frequency of bleeding at 90 days in patients with and without antidepressants. In the univariate logistic regression analysis, factors associated with new visits for gastrointestinal bleeding included chronic renal failure, treatment with antidepressants and use of antiplatelet drugs (p < 0.05) (Table 2). The multivariate model found that the presence of chronic renal failure (OR: 2.50; 95%CI 1.07–5.85; p = 0.034), treatment with antidepressants (OR: 2.56; 95%CI 1.07–6.14; p = 0.034) and the use of antiplatelet drugs (OR: 4.55; 95%CI 1.27–16.28; p = 0.020) were statistically significantly associated with an increased risk of re-consultation for gastrointestinal bleeding.

Figure 1.

Percentage of patients with emergency revisits for bleeding episodes within 90 days.

To date, few studies have investigated the effect of antidepressants on the occurrence of gastrointestinal bleeding in patients with previous hemorrhagic episodes. This study demonstrates that the combination of anticoagulants and antidepressants is associated with a significantly increased risk of revisits for gastrointestinal bleeding compared to the use of anticoagulants alone (DOACs or VKAs). These results are consistent with those reported in previous studies, in which the use of antidepressants and anticoagulants is associated with an increased risk of bleeding.14,19,20

In accordance with Giovanni et al.,7 our study demonstrates the high prevalence of antidepressant prescription in the elderly population, approaching one third of this population. Since older adults inherently carry a higher baseline risk of gastrointestinal bleeding, particularly when on anticoagulation, our findings emphasize the need for careful, individualized evaluation of antidepressant therapy in this demographic. Factors such as comorbidity, type of antidepressant, duration of treatment and bleeding risk should be carefully considered when establishing the therapeutic regimen. Although further evidence is needed to establish firm recommendations, a more personalized assessment may help to minimize complications associated with combination therapy.

On the other hand, the results of this study have found a significant relationship between gastrointestinal bleeding in anticoagulated patients with chronic kidney disease and those receiving antiplatelet drugs. The association between chronic renal disease and gastrointestinal bleeding in patients on anticoagulants has been previously published by several authors.21 However, there are discrepancies in the literature regarding these associations,22 which is why we consider it necessary to conduct new studies that take these variables into account.

These findings underscore the importance of carefully assessing the indication of antidepressants in anticoagulated patients, especially those with other risk factors for gastrointestinal bleeding. Given the high use of antidepressants in the elderly population,7 and their vulnerability to adverse events, the need for more careful and personalized prescribing is highlighted. Although the sample was limited, the findings open the door to further research, including variables such as renal function and antiplatelet use, in order to better guide clinical decision-making and reduce avoidable complications. It has been established that the risk of gastrointestinal bleeding differs among individual DOACs,6 and such variability may influence the selection of the most appropriate anticoagulant in patients treated with antidepressants in clinical practice. Unfortunately, the sample size of our study did not allow for stratified analyses by specific DOAC agent, an important aspect that should be addressed in future research.

Limitations of our study include the retrospective analysis and the absence of data on adequate patient adherence to treatment. Furthermore, it did not investigate possible neoplasia or polyps as causes of bleeding for the different groups of anticoagulants, which have previously been established as a cause of approximately 10% of gastrointestinal bleeding episodes.23 The limited sample size also restricted our ability to perform subgroup analyses comparing different classes of antidepressants. Future prospective studies, with larger sample sizes and analyses differentiated by type of antidepressant and anticoagulant, will be necessary to establish robust clinical recommendations and guide preventive interventions aimed at reducing morbidity associated with recurrent bleeding events in this vulnerable population.

In conclusion, in anticoagulated elderly patients with an episode of gastrointestinal bleeding, the concomitant use of antidepressants is associated with a significantly increased risk of revisit to the emergency department for new bleeding episodes within 90 days of hospital discharge. Likewise, the presence of chronic kidney disease and concomitant use of antiplatelet agents are independent risk factors for bleeding recurrence in this clinical context. The prescription of antidepressants in this population group is common in routine clinical practice, so a thorough risk–benefit assessment should be performed, seeking strategies to minimize risk, such as dose adjustment, review of indications and optimisation of the use of gastroprotective therapy.

All authors accept the responsibilities outlined by the International Committee of Medical Journal Editors (available at http://www.icmje.org/).

In the event of publication, the authors grant, exclusively, the rights of reproduction, distribution, translation, and public communication (by any means or support—sound, audiovisual, or electronic) of our work to Farmacia Hospitalaria and, by extension, to SEFH. A rights transfer agreement will be signed at the time of manuscript submission through the online manuscript management system.

Andrea Garrido Quintero: Writing – review & editing, Writing – original draft, Supervision, Methodology, Formal analysis, Data curation, Conceptualization. Silvia Cercas Lobo: Writing – original draft, Investigation, Data curation. Jesus Ruiz Ramos: Writing – review & editing, Supervision, Data curation, Conceptualization. Adrián Plaza Díez: Writing – original draft, Methodology, Data curation. Ane González Díez: Writing – original draft, Methodology, Data curation. Antonio Cañón Santos: Writing – original draft, Validation, Formal analysis, Conceptualization. Ana Villarejo Jiménez: Writing – original draft, Methodology, Data curation. María Carmenza Pérez Méndez: Writing – review & editing, Validation, Methodology, Formal analysis, Conceptualization. Mireia Puig Campmany: Writing – review & editing, Writing – original draft, Validation, Supervision, Conceptualization.

None.