Impact on Semen Parameters and Testicular Histopathology
Multiple studies indicate that SSRI use may negatively affect semen parameters. However, findings across studies are inconsistent. A systematic review and meta-analysis by Xu et al. demonstrated statistically significant impairments in sperm morphology, concentration, and motility among SSRI users. No effect was found on semen volume. These changes were most pronounced within the first three months of exposure, suggesting a time-dependent effect (14). Similarly, Oliveira et al. reported reductions in serum testosterone levels, decreased sperm production, diminished sperm reserves, and prolonged epididymal transit time (20). Long-term SSRI exposure (at least 6 months), particularly with agents such as citalopram and sertraline, has also been associated with reduced total sperm count and motility compared with healthy controls (6).
Prospective clinical data further support these observations. Specifically, treatment with escitalopram at a dose of 10 mg/day was associated with significant declines in sperm concentration, motility, and morphology after 12 weeks compared with baseline values (12). Similarly, fluoxetine has been highlighted as a potential contributor to reproductive toxicity, including reduced reproductive organ weight and sperm concentration (9). In addition, experimental studies have shown that citalopram exposure induces reproductive toxicity in animal models, including reduced sperm count and structural damage to testicular tissue (11).
In contrast, several large observational studies have reported no significant associations between SSRI use and semen parameters. For example, a cohort study of 8,861 men undergoing fertility evaluation found no differences in semen volume, concentration, motility, or morphology between SSRI users and non-users after adjustment for confounders (15). Likewise, a retrospective analysis of 299 men attending an infertility clinic found no significant differences between SSRI users and non-users in sperm liquefaction, motility, viscosity, or sperm count (21). Furthermore, available evidence suggests that SSRI-associated changes in semen parameters may be reversible following discontinuation of treatment, indicating a potential transient effect (9). Overall, these inconsistencies highlight the variability of clinical findings and suggest that SSRI-associated changes in semen parameters may not be uniform across all populations. Additionally, the potential confounding effect of underlying psychiatric conditions has also been emphasized.
In an experimental model of chronic stress–induced depression, untreated stress was associated with marked testicular damage and elevated inflammatory markers. In contrast, treatment with fluoxetine or vortioxetine improved testicular histopathology and reduced inflammatory and apoptotic markers compared to untreated stressed animals (17). These findings suggest that the reproductive effects observed with SSRI use may, in part, reflect the influence of depression or chronic stress, rather than a direct toxic effect of the medication itself.
Molecular Mechanisms: DNA Integrity, Oxidative Stress, and Ion Channels
Research indicates that SSRIs may impair sperm function at a molecular level, even when routine semen parameters appear normal. Studies involving sertraline and paroxetine have demonstrated significantly higher rates of DNA fragmentation compared with behavioral therapy or untreated controls (6). A study on healthy volunteers treated with paroxetine revealed a significant increase in sperm DNA fragmentation (from 13.8% to 30.3%). This increase occurred despite normal standard semen analysis, suggesting subclinical reproductive toxicity (16). These findings are supported by meta-analytic data and narrative reviews. These reviews indicate a significant increase in the sperm DNA fragmentation index among SSRI users (p = 0.0002) (9,14).
Two principal mechanisms have been proposed to explain these molecular effects: oxidative stress and ion channel inhibition. In vivo studies in male rats have shown that citalopram exposure reduces testicular glutathione levels and increases oxidative stress, accompanied by sperm DNA damage and histopathological alterations (11). In parallel, in vitro studies using human sperm exposed to fluoxetine have demonstrated elevated levels of ROS and MDA. These findings also include a reduction in TAC and activation of apoptotic pathways, shown by upregulation of Caspase-8, Caspase-9, and Bax, along with downregulation of the anti-apoptotic BCL-2 gene (13).
In addition to oxidative mechanisms, inhibition of sperm-specific ion channels has also emerged as a relevant pathway. For example, several SSRIs, particularly sertraline, have been identified as potent inhibitors of the CatSper calcium channel. Given that CatSper-mediated calcium influx is essential for sperm hyperactivation and the acrosome reaction, its inhibition may impair fertilization capacity independently of standard semen parameters (10).
Pregnancy and Neonatal Outcomes
Epidemiological data from large population-based cohorts are generally reassuring regarding the safety of paternal SSRI use in relation to pregnancy and neonatal outcomes (22,23). For instance, a nationwide cohort study including 13,547 children exposed to paternal SSRI use during the three months prior to conception found no significant associations with SGA birth, low Apgar scores, major congenital malformations, or infection risk (3). However, a modest but statistically significant increase in the risk of preterm birth was observed (OR: 1.15). This was particularly associated with paternal citalopram and escitalopram use (3). Nevertheless, these findings should be interpreted with caution due to potential confounding by indication. The underlying paternal psychiatric condition could independently influence pregnancy outcomes.
Regarding long-term neurodevelopmental risks, studies have largely pointed towards non-causal associations. Yang et al. investigated the relationship between paternal SSRI use and ADHD and found a statistically significant increase in risk. However, a similar risk was observed in children whose fathers had discontinued treatment long before conception. These findings suggest that the association stems from confounding factors, such as the genetic transmission of psychopathology or shared environmental factors, rather than direct pharmacological effects (18). Similarly, in an analysis of ASD, Yang et al. reported that while initial data suggested an increased risk, further stratification revealed this was likely driven by confounding by indication. The risk persisted among offspring of ‘former users’. Limitations of this study included reliance on dispensing data and the inability to adjust for diagnoses managed exclusively in primary care settings (19). In a similar vein, another large nationwide cohort study by Viktorin et al. examined the impact of paternal antidepressant use on preterm birth, malformations, ASD, and intellectual disability. This study found no significant associations and concluded that paternal use is generally safe (2).
Several methodological limitations should be considered when interpreting the findings of this review. First, the reviewed literature is partly based on animal studies, primarily involving rats and mice. The drug doses used in these experiments are often higher than therapeutic levels, and physiological differences between species may limit their relevance to human clinical settings. Second, most large and high-quality epidemiological data originate from Nordic national registries, particularly those of Sweden and Denmark. This geographic clustering may limit the external validity and generalizability of the results to populations with different ethnic, genetic, and healthcare characteristics. Third, because of confounding by indication, it is difficult to determine whether the reported neonatal or neurodevelopmental outcomes result from SSRI exposure itself or from the underlying paternal psychiatric disorder. Prescription-based analyses cannot verify actual medication use or exposure timing, and missing data on key lifestyle factors such as smoking, alcohol intake, and body mass index may have contributed to residual confounding. Finally, the limited clinical evidence available for newer agents, such as vortioxetine and vilazodone, combined with small sample sizes in some studies, makes it difficult to draw firm conclusions about the reproductive safety of the entire SSRI class.
DISCUSSION
Impact on Semen Parameters and Testicular Histopathology
Multiple studies indicate that SSRI use may negatively affect semen parameters. However, findings across studies are inconsistent. A systematic review and meta-analysis by Xu et al. demonstrated statistically significant impairments in sperm morphology, concentration, and motility among SSRI users. No effect was found on semen volume. These changes were most pronounced within the first three months of exposure, suggesting a time-dependent effect (14). Similarly, Oliveira et al. reported reductions in serum testosterone levels, decreased sperm production, diminished sperm reserves, and prolonged epididymal transit time (20). Long-term SSRI exposure (at least 6 months), particularly with agents such as citalopram and sertraline, has also been associated with reduced total sperm count and motility compared with healthy controls (6).
Prospective clinical data further support these observations. Specifically, treatment with escitalopram at a dose of 10 mg/day was associated with significant declines in sperm concentration, motility, and morphology after 12 weeks compared with baseline values (12). Similarly, fluoxetine has been highlighted as a potential contributor to reproductive toxicity, including reduced reproductive organ weight and sperm concentration (9). In addition, experimental studies have shown that citalopram exposure induces reproductive toxicity in animal models, including reduced sperm count and structural damage to testicular tissue (11).
In contrast, several large observational studies have reported no significant associations between SSRI use and semen parameters. For example, a cohort study of 8,861 men undergoing fertility evaluation found no differences in semen volume, concentration, motility, or morphology between SSRI users and non-users after adjustment for confounders (15). Likewise, a retrospective analysis of 299 men attending an infertility clinic found no significant differences between SSRI users and non-users in sperm liquefaction, motility, viscosity, or sperm count (21). Furthermore, available evidence suggests that SSRI-associated changes in semen parameters may be reversible following discontinuation of treatment, indicating a potential transient effect (9). Overall, these inconsistencies highlight the variability of clinical findings and suggest that SSRI-associated changes in semen parameters may not be uniform across all populations. Additionally, the potential confounding effect of underlying psychiatric conditions has also been emphasized.
In an experimental model of chronic stress–induced depression, untreated stress was associated with marked testicular damage and elevated inflammatory markers. In contrast, treatment with fluoxetine or vortioxetine improved testicular histopathology and reduced inflammatory and apoptotic markers compared to untreated stressed animals (17). These findings suggest that the reproductive effects observed with SSRI use may, in part, reflect the influence of depression or chronic stress, rather than a direct toxic effect of the medication itself.
Molecular Mechanisms: DNA Integrity, Oxidative Stress, and Ion Channels
Research indicates that SSRIs may impair sperm function at a molecular level, even when routine semen parameters appear normal. Studies involving sertraline and paroxetine have demonstrated significantly higher rates of DNA fragmentation compared with behavioral therapy or untreated controls (6). A study on healthy volunteers treated with paroxetine revealed a significant increase in sperm DNA fragmentation (from 13.8% to 30.3%). This increase occurred despite normal standard semen analysis, suggesting subclinical reproductive toxicity (16). These findings are supported by meta-analytic data and narrative reviews. These reviews indicate a significant increase in the sperm DNA fragmentation index among SSRI users (p = 0.0002) (9,14).
Two principal mechanisms have been proposed to explain these molecular effects: oxidative stress and ion channel inhibition. In vivo studies in male rats have shown that citalopram exposure reduces testicular glutathione levels and increases oxidative stress, accompanied by sperm DNA damage and histopathological alterations (11). In parallel, in vitro studies using human sperm exposed to fluoxetine have demonstrated elevated levels of ROS and MDA. These findings also include a reduction in TAC and activation of apoptotic pathways, shown by upregulation of Caspase-8, Caspase-9, and Bax, along with downregulation of the anti-apoptotic BCL-2 gene (13).
In addition to oxidative mechanisms, inhibition of sperm-specific ion channels has also emerged as a relevant pathway. For example, several SSRIs, particularly sertraline, have been identified as potent inhibitors of the CatSper calcium channel. Given that CatSper-mediated calcium influx is essential for sperm hyperactivation and the acrosome reaction, its inhibition may impair fertilization capacity independently of standard semen parameters (10).
Pregnancy and Neonatal Outcomes
Epidemiological data from large population-based cohorts are generally reassuring regarding the safety of paternal SSRI use in relation to pregnancy and neonatal outcomes (22,23). For instance, a nationwide cohort study including 13,547 children exposed to paternal SSRI use during the three months prior to conception found no significant associations with SGA birth, low Apgar scores, major congenital malformations, or infection risk (3). However, a modest but statistically significant increase in the risk of preterm birth was observed (OR: 1.15). This was particularly associated with paternal citalopram and escitalopram use (3). Nevertheless, these findings should be interpreted with caution due to potential confounding by indication. The underlying paternal psychiatric condition could independently influence pregnancy outcomes.
Regarding long-term neurodevelopmental risks, studies have largely pointed towards non-causal associations. Yang et al. investigated the relationship between paternal SSRI use and ADHD and found a statistically significant increase in risk. However, a similar risk was observed in children whose fathers had discontinued treatment long before conception. These findings suggest that the association stems from confounding factors, such as the genetic transmission of psychopathology or shared environmental factors, rather than direct pharmacological effects (18). Similarly, in an analysis of ASD, Yang et al. reported that while initial data suggested an increased risk, further stratification revealed this was likely driven by confounding by indication. The risk persisted among offspring of ‘former users’. Limitations of this study included reliance on dispensing data and the inability to adjust for diagnoses managed exclusively in primary care settings (19). In a similar vein, another large nationwide cohort study by Viktorin et al. examined the impact of paternal antidepressant use on preterm birth, malformations, ASD, and intellectual disability. This study found no significant associations and concluded that paternal use is generally safe (2).
Several methodological limitations should be considered when interpreting the findings of this review. First, the reviewed literature is partly based on animal studies, primarily involving rats and mice. The drug doses used in these experiments are often higher than therapeutic levels, and physiological differences between species may limit their relevance to human clinical settings. Second, most large and high-quality epidemiological data originate from Nordic national registries, particularly those of Sweden and Denmark. This geographic clustering may limit the external validity and generalizability of the results to populations with different ethnic, genetic, and healthcare characteristics. Third, because of confounding by indication, it is difficult to determine whether the reported neonatal or neurodevelopmental outcomes result from SSRI exposure itself or from the underlying paternal psychiatric disorder. Prescription-based analyses cannot verify actual medication use or exposure timing, and missing data on key lifestyle factors such as smoking, alcohol intake, and body mass index may have contributed to residual confounding. Finally, the limited clinical evidence available for newer agents, such as vortioxetine and vilazodone, combined with small sample sizes in some studies, makes it difficult to draw firm conclusions about the reproductive safety of the entire SSRI class.