Analysis of the Use of Duloxetine in the High-dose Range (90 to 120 mg) in the Psychiatric Clinic for the Treatment of Patients With Major Depressive Disorder

Abstract

Background and Objectives:

Duloxetine is a serotonin and norepinephrine reuptake inhibitor antidepressant drug. The goal of this project was to analyze the characteristics of patients with major depressive disorder (MDD) treated with duloxetine in the high-dose range (90 to 120 mg).

Methods:

A retrospective, multicenter study was designed to analyze 20 indicators on the clinical use of duloxetine (90 to 120 mg) in adults with MDD who were referred to the psychiatric clinic.

Results:

A total of 175 psychiatrists participated in the study and collected information on 1050 patients with a mean age of 51.3 years (62.9% women). A total of 55.4% had comorbid psychiatric conditions, with dysthymia (17.2%), generalized anxiety disorder (15%), and personality disorders (13.1%) being the most common. A total of 84.7% had nonpsychiatric conditions, including chronic pain (28.8%), fibromyalgia (22.4%), sleep disorders (20.5%), arterial hypertension (17.8%), dyslipidemia (16.3%), neuropathic pain (11.4%), and cognitive impairment (5.8%). In all, 45.3% of the sample were found to be markedly to extremely ill. A total of 84.3% had received 1 to more than 4 prior lines of therapy. The most common reason for prescribing duloxetine in the high-dose range was a lack of efficacy of previous treatment (41%) or because the patient showed only a partial response to lower doses of duloxetine (30.7%).

Conclusion:

A greater than expected proportion of patients with advanced, marked, or severe disease, psychiatric comorbidity, sleep disorders, neuropathic pain, and cognitive impairment was observed among the subjects with MDD treated with high-dose duloxetine.

Major depressive disorder (MDD) is a relevant health problem due to its high prevalence, the disability it causes, the worsening of patients' quality of life, and its huge economic impact. In Europe, the annual prevalence in individuals 18 to 65 years of age is estimated to be 6.1%, making it the most frequent mental disorder.¹ In Spain, the prevalence of depression in primary care is reported to be 4.1%,² although that figure is likely underestimated since ~28% of patients with MDD are undiagnosed.³

In our health care setting, primary care centers are the first point of access to the health care system for patients with depression, and most of those patients continue their care in these centers. However, some patients require referral to specialized psychiatric care, particularly in situations of inadequate response to treatment, frequent relapses, presence of suicidal ideation or psychotic symptoms, or presence of other psychiatric conditions or concomitant diseases that impede or complicate the use of antidepressants.⁴

Although effective treatments are available, only 34% of patients with MDD attending primary care are treated with antidepressants, and there is also great variability in the management of these patients.³ The association of depression with other medical and psychiatric conditions is very common and is related to the prognosis of MDD and to suicidal ideation. The presence of comorbidity may also influence the choice of the most appropriate treatments for MDD based on the safety profile of the antidepressant, its interactions with treatments for physical or other psychiatric disorders, or the benefits that the drug may provide in relation to the comorbid conditions.⁴

Duloxetine is a serotonin and norepinephrine reuptake inhibitor (SNRI) antidepressant initially approved for the treatment of MDD,⁵ and currently also indicated by the European Medicines Agency (EMA) and the US Food and Drug Administration (FDA) for the treatment of generalized anxiety disorder and diabetic peripheral neuropathic pain.^6–8 Additional approved indications in Europe include stress urinary incontinence, and in the United States include fibromyalgia and chronic musculoskeletal pain. Due to its pharmacological properties, duloxetine has been studied in the treatment of different psychiatric disorders and may represent a valuable option for patients with other disorders coexisting with MDD and generalized anxiety disorder.

The guidelines of the British Association for Psychopharmacology recommend the use of high doses of antidepressants if the expected side effects are minimal, if there has been improvement with the treatment but a satisfactory response has not yet been achieved, and if the drug has a dose-response effect—since maintaining an ineffective low dose does not benefit the patient, while a high dose is related to a higher recovery rate.⁹ However, it is currently not clear in which patients the specialist may decide to use duloxetine in the high-dose range of 90 to 120 mg, and whether its use could be related to previous resistance to treatments or to specific clinical situations.

Therefore, the goal of this study was to analyze the clinical use of high-dose duloxetine in patients with MDD treated in psychiatric practice. Specifically, the study sought to address the following issues related to the use of high-dose duloxetine:

1. The demographic and clinical characteristics of patients receiving duloxetine at doses of 90 to 120 mg/day.

2. The psychiatric and nonpsychiatric comorbidities associated with its use.

3. Prior treatment patterns and levels of illness severity characterizing patients receiving duloxetine at doses of 90 to 120 mg/day.

4. Clinical reasons leading psychiatrists to prescribe duloxetine within the high-dose range in real-world practice.

METHODS

Study Design and Ethical Norms

A retrospective, descriptive drug utilization study with aggregated data was designed. The study was approved by the Research Ethics Committee of Hospital Universitario Puerta de Hierro (Majadahonda, Madrid, Spain) on July 4, 2022. The study was carried out following the guidelines of the International Society for Pharmacoepidemiology (ISPE) with reference to Good Epidemiological Practice and the basic ethical principles of the Declaration of Helsinki. Due to the study design involved (retrospective data collection and data aggregation), the need to obtain patient-informed consent was waived. A total of 175 psychiatry specialists from 37 Spanish provinces participated in the study, which was conducted between September 2022 and January 2023.

Seven structural indicators and 20 process indicators were selected. Structural indicators included participating physician sex and age, years of professional experience in psychiatry, type of center, type of practice setting, monthly number of patients with MDD, and monthly number of new patients with MDD. Process indicators included patient sex, mean age, age group, presence of psychiatric comorbidities, types of psychiatric comorbidities, nonpsychiatric comorbidities, presence of somatic symptoms, types of somatic symptoms, current or recent substance use, first-degree family history, lifetime suicide attempts, Clinical Global Impression-Severity score, number of previous treatment lines, antidepressant treatment before psychiatric consultation, use of an SSRI as first-line treatment, treatment at the time of psychiatric consultation, action taken regarding duloxetine treatment, reasons for prescribing duloxetine at 90 or 120 mg/day, consideration of smoking-related CYP1A2 induction, and treatments combined with high-dose duloxetine.

The process indicators were evaluated by retrospectively collecting information from 6 consecutive medical records of patients who were referred to the psychiatry clinic. The source of information was the medical record, and the patients were required to meet the following inclusion criteria: (1) to be older than 18 years of age; (2) to meet the diagnostic criteria for MDD (DSM-5); (3) to be under treatment with duloxetine at doses of 90 to 120 mg before the consultation or after the in-office evaluation; and (4) to be an outpatient. According to the European Summary of Product Characteristics (EMA/AEMPS), duloxetine is approved for the treatment of MDD and diabetic peripheral neuropathic pain at doses of 60 to 120 mg/day, and for generalized anxiety disorder at doses ranging from 30 to 120 mg/day, with 60 mg/day being considered the usual recommended maintenance dose.⁶

The participating physicians received a form on which to collect the data from the medical records. For each indicator, the participating physicians specified whether the criteria were met for each patient or whether measurement of the indicator was not applicable to the patient involved. The data were reported on an aggregated basis on an electronic platform with restricted access.

Statistical Analysis

Statistical analyses were performed, with the participating center considered the unit of analysis. Each investigator provided aggregated information derived from 6 eligible patients. For process and outcome indicators, physicians reported the number of patients fulfilling each indicator and the number of valid cases. Proportions were calculated at the center level based on the number of valid cases for each indicator.

Descriptive analyses included calculation of frequencies for qualitative variables and the mean, standard deviation (SD), minimum and maximum values, and 95% confidence intervals (95% CIs) for quantitative variables. Comparisons were made using the Fisher exact test or the χ² test in the case of qualitative variables, and the Student t test for quantitative variables, or the applicable nonparametric tests in the absence of normal data distribution. Statistical significance was considered to be P<0.05. The SPSS version 27.0 statistical package was used for analysis (IBM Corp., Armonk, NY).

Sample Size Calculation

The analytical unit was the consulting office of the participating physician. With the sample of 175 investigators, a statistical power of 89% was estimated for the evaluation of the selected process indicators, assuming a proportion of 50%, which represents the most conservative scenario and results in the maximum standard error for proportion estimates. This provided an estimated accuracy of ±7% for the corresponding confidence intervals for proportions, determined by normal approximation with a 2-sided alpha error of 0.05 (Sample Power IBM-SPSS).

RESULTS

Aggregated information was obtained from 1050 patients with MDD who received duloxetine in the daily dose range of 90 to 120 mg. Most of the psychiatrists’ clinics were public (90.3%) and part of a mental health institution (86.3%), whereas 7.4% and 6.3% were hospital and private consultations, respectively. Twenty-two percent of the patients seen in the psychiatry clinics were new cases of individuals with MDD.

The results based on demographic data and medical records are shown in Table 1. Table 2 compares the prevalence of comorbid nonpsychiatric and psychiatric conditions among the patients treated with duloxetine at doses of 90 to 120 mg with the population of patients with depression and with the general population. Compared with previously reported data in patients with MDD, patients receiving duloxetine at doses of 90 to 120 mg showed significantly higher prevalences of sleep disorders (20.5%) and cognitive impairment (5.8%). In contrast, lower prevalences were observed for most other explored conditions compared with prevalences reported in data on patients with MDD. Among nonpsychiatric comorbidities, liver failure and renal failure were, respectively, 31.3 and 6.1 times less frequent than expected, while cerebrovascular disease and movement disorders were 3.5 and 3.6 times less frequent, respectively. Among the psychiatric comorbidities, schizophrenia was 11.8 times less frequent, panic disorder 7.2 times less frequent, and obsessive-compulsive disorder 5.7 times less frequent than reported in data on patients with MDD. Similar prevalences were observed for arterial hypertension, diabetes, sexual dysfunction, epilepsy, dysthymia, and personality disorders.

TABLE 1.

Demographic and Clinical Data of the Patients With Major Depressive Disorder Treated With Duloxetine 90 to 120 mg

	% (95% CI) N=1050*
Sex
Female	62.9 (59.8-65.9)
Male	37 (33.9-40)
Age	51.3 (50.2-52.4)
History in first-degree relatives
Psychiatric disease	25.1 (21.5-28.7)
Affective disorders	29 (25.6-32.5)
Suicide	4.4 (2.9-5.9)
Nonpsychiatric comorbidity	84.7 (81.9-87.4)
Psychiatric comorbidity	55.4 (51.2-59.6)
Somatization disorder	63.5 (59.7-67.4)
Type of somatization
Pain	42.1 (37.9-46.3)
Gastrointestinal	14.9 (12.1-17.6)
Neurological	10.1 (7.5-12.7)
Cardiopulmonary	6.9 (4.9-8.9)
Other type	4.7 (2.6-6.7)
Suicide attempt	20.1 (17.1-23.1)
Substance abuse
None	61.1 (57-65.1)
Smoking	29.3 (25.8-32.8)
Alcohol	13.2 (10.9-15.4)
Other substances	7.6 (5.5-9.8)

^*Values are presented as mean (95% CI) based on aggregated data reported by participating investigators (N=175), each of whom contributed information from 6 patients.

TABLE 2.

Prevalence of Physical and Psychiatric Diseases in Patients With Major Depressive Disorder Selected for Treatment With Duloxetine at Doses of 90 to 120 mg Versus Patients With Depression and the General Population

	Patients with Major Depressive Disorder Receiving Duloxetine 90-120 mg % (95% CI) N=1050*	Ratio to Prevalence in Patients With Major Depressive Disorder†	Patients With Major Depressive Disorder Prevalence (%)	General Population Prevalence (%)
Nonpsychiatric comorbidities by order of frequency in the study
Chronic pain	28.8 (24.9-32.6)	<2.3	6510	15-2511
Fibromyalgia	22.4 (19.2-25.6)	<1.4	29.3-31.212	2.3713
Sleep disturbances	20.5 (16.4-24.5)	> 2.8	7.413	5.42 6.0714
Hypertension	17.8 (15.2-20.4)	<2.7=1.2	47.712 20.715	19.816
Dyslipidemia	16.3 (13.6-18.9)	<2.5	40.4-40.611	17.916
Neuropathic pain	11.4 (8.8-14.1)	—	—	5-817
Diabetes mellitus	10 (8-12)	<1.9=1.2	18.3-1912 11.715	7.816
Obesity (body mass index >30 kg/m2)	9.4 (7.1-11.7)	<2.3	19.4-21.212	1614 17.416
Other disease	9.3 (6.5-12.1)	—	—	—
Sexual dysfunction	7.4 (5.3-9.6)	=1.7	9.10-12.912	50% at some point in life18
Cardiovascular disease	7.3 (5.3-9.4)	<1.8	13.2-12.912	2.0914 2.0916
Cognitive impairment	5.8 (4-7.6)	>1.9	2.03-3.0412	3.22
Urinary incontinence	3.3 (1.6-5)	—	—	22.6 in females/9.5 in males (same age group)19
Cerebrovascular diseases	3.2 (1.9-4.6)	<3.5	11.1-11.212	1.5114 1.5116
Movement disorders	2.1 (1-3.1)	<3.6	5.67-7.4712	<120
Epilepsy	1.5 (0.7-2.4)	=1.3	212	0.821
Chronic renal failure	1.4 (0.6-2.3)	<6.1	8.82-8.4712	17.4 (same age group)22
Coagulation disorders	1.1 (0.1-2.2)	<2.6	3.04-2.8912	—
Chronic liver failure	0.4 (0-0.8)	<31.3	11.3-12.512	1.523
Psychiatric comorbidities by order of frequency in the study
Somatization disorder	63.5 (59.7-67.4)	=1	52.4-65.512	0.552
Dysthymic disorder	17.2 (14.3-20.1)	=0.9=0.9	15.724 16.2125	1.526
Generalized anxiety disorder	15 (12.4-17.5)	<3.7 <3.3 <4.3	55.224 50 (44.5-56.6)15 60-6512	3.126 6.72
Personality disorders	13.1 (10.7-15.6)	=1 <3.8 <2.1	6.49-14.312 30-5027 2825	0.672
Alcohol use disorder	7 (5.3-8.8)	<2 <2.8	1428 19.2925	3.126
Social phobia	6.5 (4.7-8.2)	<1.6	10.4925	6.826 0.22
Panic disorder	4.7 (3.1-6.2)	<7.2 <2.5	33.824 11.5625	2.726
Other substance use disorder (other than alcohol or nicotine)	4.6 (2.9-6.3)	<4.4=1 <3.2	12.5-20.1112 4.628 14.7425	3.826
Other disorders	3.3 (1.8-4.9)	—	—	—
Obsessive-compulsive disorder	2.6 (1.7-3.5)	<5.7	14.8225	8.926 0.112
Eating disorders	2.5 (1.5-3.5)	—	—	0.142
Other psychoses	2.3 (1.2-3.3)	—	—	0.192
Posttraumatic stress disorder	1.5 (0.8-2.2)	—	—	3.526
Attention-deficit/hyperactivity disorder	1.1 (0.4-1.9)	<2.9	2.35-3.2012	4.126 1.82
Schizophrenia	0.6 (0.1-1)	<11.8	2.39-7.0712	0.372

^*Percentages are presented as mean estimates (95% CI) based on aggregated data reported by participating investigators (N=175), each of whom contributed information from 6 patients.

^†Ratio of prevalence in patients with major depressive disorder divided by prevalence in the patients in the study: <statistically significant lesser proportion; >statistically significant greater proportion; =no statistically significant differences in the proportion. Significance was determined by comparing the prevalence in patients with major depressive disorder with the 95% confidence interval of the prevalence observed in the study sample. A prevalence above the upper limit of the 95% CI was considered significantly higher, a prevalence below the lower limit was considered significantly lower, and a prevalence within the 95% CI was considered not significantly different.

In terms of the severity of depression, the patients were classified using the Clinical Global Impression scale reported by the psychiatrist, as shown in Figure 1. The results showed that ratings for 45.3% of patients corresponded to markedly, severely or extremely ill.

FIGURE 1.

Severity of major depressive disorder according to the Clinical Global Impression (CGI) scale. Patients were classified using the CGI scale reported by the psychiatrist. Blue bars represent patients classified as normal to moderately ill, and red bars represent those classified as markedly to extremely ill. The proportion of patients in each CGI severity grade at the time of consultation is shown (N=1050).

In terms of previous medication therapy, 15.7% of the patients (95% CI: 12.6-18.8) had not received any previous therapy, 27.7% (95% CI: 24.3-31.1) had received 1 previous line of therapy, 26.6% (95% CI: 23.2-29.9) 2 lines of therapy, 20% (95% CI: 16.3-23.7) 3 lines of therapy, and 10% 4 or more lines of therapy. A total of 80.4% of the patients (95% CI: 76.4-84.5) who had received an antidepressant had been treated with a selective serotonin reuptake inhibitor (SSRI) as the first line of therapy.

Table 3 shows the proportion of patients receiving each treatment for the current major depressive episode before versus after the psychiatric consultation. A total of 17.2% of the patients (95% CI: 13.9-20.6) were not receiving any treatment for the current major depressive episode before psychiatric consultation. In 28.1% of the patients (95% CI: 23.6-32.6), no drug was combined with duloxetine 90 to 120 mg for the treatment of depression or psychiatric comorbidities after psychiatric consultation.

TABLE 3.

Treatments Administered to the Patient for the Current Major Depressive Episode and Psychiatric Comorbidities, Before and After Psychiatric Consultation

Active Substance or Class of Drugs by Order of Frequency of Administration Before Psychiatric Consultation	Before Psychiatric Consultation % (95% CI) N=1050*	After Psychiatric Consultation % (95% CI) N=1050*	Difference†
Benzodiazepine	50.8 (45.3-56.2)	46.8 (41.5-52.1)	=4
Duloxetine	38.9 (30.3-47.4)	100‡	▲61.1
Escitalopram	15.9 (13.5-18.3)	1.4 (0.6-2.3)	▼14.5
Mirtazapine	14.3 (12-16.6)	17.1 (14.5-19.8)	▲2.8
Sertraline	13.9 (11.5-16.3)	1.2 (0.5-2)	▼12.7
Paroxetine	10.7 (8.8-12.5)	1 (0.3-1.6)	▼9.7
Trazodone	10.3 (8-12.6)	14.3 (11.7-16.9)	▲4
Fluoxetine	9 (7.1-11)	0.6 (0-1.1)	▼8.4
Quetiapine	8 (6-10)	10.3 (8-12.5)	=2.3
Venlafaxine	7.9 (6-9.8)	0.8 (0.2-1.3)	▼7.1
Citalopram	6.8 (4.9-8.6)	0.6 (0-1.1)	▼6.2
Amitriptyline	6.4 (4.5-8.2)	4.9 (3.2-6.5)	=1.5
Desvenlafaxine	6 (4.3-7.7)	1 (0.1-1.8)	▼5
Vortioxetine	4.2 (2.7-5.7)	1.3 (0.6-2.1)	▼2.9
Other euthymia-inducing drugs	3.7 (2.2-5.2)	4.6 (2.8-6.4)	=0.9
Bupropion	2.7 (1.6-3.7)	2.9 (1.7-4)	=0.2
Aripiprazole	2.7 (1.6-3.8)	5.2 (3.8-6.7)	▲2.5
Olanzapine	2.4 (1.5-3.3)	3.7 (2.6-4.8)	▲1.3
Agomelatine	1.6 (0.8-2.4)	2.1 (1.1-3)	=0.5
Lithium salts	1.4 (0.4-2.4)	1.1 (0.3-2)	=0.3
Clomipramine	1.1 (0.5-1.8)	0.7 (0.2-1.2)	=0.4
Other atypical antipsychotics	1.1 (0.5-1.8)	1.3 (0.6-2.1)	=0.2
Fluvoxamine	1 (0.4-1.7)	0.1 (−0.1, 0.3)	▼0.9
Risperidone	1 (0.3-1.6)	0.6 (0.1-1)	=0.4
Typical antipsychotics	1 (0.3-1.6)	0.9 (0.2-1.5)	=0.1
Mianserin	0.9 (0.2-1.5)	1 (0.4-1.7)	=0.1
Other enhancers	0.9 (0-1.7)	1 (0.2-1.7)	=0.1
Reboxetine	0.7 (0.1-1.2)	0.7 (0.1-1.2)	=0
Paliperidone	0.7 (0.2-1.2)	0.6 (0-1.1)	=0.1
Methylphenidate or other psychostimulants	0.6 (0-1.1)	0.6 (0-1.1)	=0
Nortriptyline	0.2 (−0.1, 0.5)	0	=0.2
Imipramine	0.1 (−0.1, 0.3)	0	=0.1

^*Percentages are presented as mean estimates (95% CI) based on aggregated data reported by participating investigators (N=175), each of whom contributed information from 6 patients.

^†Difference in the proportion of patients receiving each treatment before versus after psychiatric consultation: =nonsignificant differences; ▲significant increase in the proportion of use of the treatment (P<0.05); ▼significant decrease in the proportion of use of the treatment (P<0.05).

^‡An inclusion criterion of this study was that patients had to be treated with high doses of duloxetine after the psychiatric consultation.

Regarding the patients who were smokers, in 35.3% of the cases, it was taken into account that plasma duloxetine concentrations are almost 50% lower in smokers versus nonsmokers, due to the fact that tobacco acts as an inducer of isoenzyme CYP1A2, and for this reason, the 90 to 120 mg dose range of duloxetine was prescribed.

The psychiatrists were asked to describe their main reason for prescribing duloxetine in the 90 to 120 mg dose range. The most common reason described was a lack of efficacy of previous treatment (41% of the patients), followed by a partial response to the 60 mg dose (26.5% of patients), followed by the prescribing psychiatrist’s usual practice of using duloxetine at doses of 90 to 120 mg (13.1% of patients). The previous treatment was modified in 11.3% of the patients due to lack of efficacy and concurrent tolerability problems. In 5.3% of the patients, the reason for prescribing duloxetine was tolerability problems with the previous treatment. Duloxetine was prescribed at high doses in 4.2% of the cases because the patient showed a partial response to the 30 mg dose of the drug.

Figure 2 shows the distribution of patients in terms of how they reached a dose of 90 to 120 mg of duloxetine after the psychiatric consultation.

FIGURE 2.

Distribution of patients regarding how they reached a dose of 90 to 120 mg of duloxetine in the psychiatric consultation. The proportion of patients in each category is shown (N=1050).

DISCUSSION

This study analyzed the clinical use of duloxetine in the high-dose range in patients with MDD, focusing on associated comorbidities and the clinical situations in which the drug is prescribed in routine psychiatric practice.

Clinical Characteristics of the Patients Receiving High-dose Duloxetine

The study included a large sample of 1050 patients with MDD receiving duloxetine in the high-dose range, with the participation of 175 physicians. The demographic characteristics were consistent with those reported in previous studies, with women representing two-thirds of the patients with MDD.^12,29 The proportion of patients with nonpsychiatric comorbidities (84.7%) was high, as reported in other studies where a 2-fold greater prevalence was observed in patients with depression.¹² The proportion of patients with psychiatric comorbidities was significantly greater than expected in our setting (45.7%),²⁴ although lower than in studies in the United States (75.5%)²⁵ and Germany (64% to 78%).¹² The frequency of a somatization disorder coincided with the data found in the literature (52.4% to 65.5% according to the severity of depression).¹² A relevant proportion of patients had a history of suicide attempt (20.1%), consistent with the increased risk of suicidal behavior described in individuals with depression.¹ The prevalence of smoking in our study appeared lower than reported in previous studies in Spanish patients with MDD.³⁰ As noted above, pharmacokinetic studies have shown that smokers exhibit reduced plasma concentrations of duloxetine due to CYP1A2 induction associated with tobacco exposure, which should be taken into account when selecting higher doses in clinical practice.^5,30,31

Role of Psychiatric and Nonpsychiatric Comorbidities

Comparison with previously published data showed that the proportion of patients with sleep disorders (20.5%) and cognitive impairment (5.8%) was significantly higher than previously reported in patients with MDD.¹² In contrast, most other explored psychiatric and nonpsychiatric comorbidities presented significantly lower prevalences, with the exception of arterial hypertension, diabetes, sexual dysfunction, epilepsy, dysthymia and personality disorders.

These findings suggest that increased use of high-dose duloxetine could be related to the presence of sleep disorders, neuropathic pain, and cognitive impairment, while its use may be lower in patients with comorbidities that were less frequent than expected in individuals with MDD, such as liver and renal failure among the physical disorders, and schizophrenia, panic disorder, and obsessive-compulsive disorders among the psychiatric comorbidities. As duloxetine is contraindicated in liver disease and severe renal disease, the less frequent use of high doses in these conditions may be justified. It is notable that the proportion of patients with comorbid generalized anxiety disorder was lower than described in other studies in patients with MDD, since duloxetine is an established first-line treatment option for generalized anxiety disorder according to various systematic reviews and clinical practice guidelines.^7,8 It is also noteworthy that the proportion of patients with chronic pain was not higher in our study, despite pain being known to increase the severity and impact of depression.¹

In patients with MDD seen in primary care in Spain, 45.7% have been found to have psychiatric comorbidity, with generalized anxiety disorder being the most common problem, presenting an odds ratio (OR) of 7.5 compared with patients without depression, followed by panic disorder (OR: 18), dysthymia (OR: 6.1), and somatization disorder (OR: 51.4). It has been reported that depression is more severe and has a greater impact upon function and quality of life in patients with comorbidities.²⁴ Similar data were also recorded in psychiatry clinics, although the comorbidity rates are usually higher in primary care settings.²⁴ The presence of psychiatric comorbidity in patients with MDD complicates clinical management and leads to a poorer prognosis with a higher probability of treatment resistance and recurrence, as well as a higher risk of suicide.²⁶ For these reasons, it is important to assess psychiatric comorbidity and individualize patient treatment in relation to the presence of other physical disease conditions.

Previous Treatment Patterns and Severity of Illness

The analysis of treatment patterns before and after psychiatric consultation showed a large proportion of patients receiving benzodiazepines, with little change in their prescription after the visit. These findings are consistent with those reported in other studies (45.1%),²⁹ and they could be explained by the presence of generalized anxiety, insomnia, or the need for muscle relaxants in patients with pain. The drugs whose use was reduced the most after the psychiatric consultation were escitalopram, sertraline, paroxetine, and fluoxetine—all of them SSRIs—due to lack of efficacy or safety issues. Considering that more than 80% of the patients in this study had been previously treated with an SSRI, this suggests that SSRIs are still being used as first-line therapy for MDD in many cases, despite current recommendations. SSRIs do not necessarily have to be the first-choice antidepressants, and treatment should be individualized based on the predominant symptoms, comorbidities, side effects, drug interactions, and response to previous treatments.⁷

In addition, a substantial proportion of patients were classified as markedly to extremely ill according to the Clinical Global Impression scale, reflecting a population with high severity of illness. This finding is consistent with the frequent need for treatment optimization or switching strategies observed in clinical practice, and it may explain the use of duloxetine within the high-dose range.

The efficacy and safety of duloxetine are currently well established. Compared with 20 antidepressant drugs, duloxetine was found to be among the 3 most effective active substances in the treatment of MDD in adults.³¹ A recent systematic review including patients with MDD, generalized anxiety disorder, neuropathic pain, and fibromyalgia confirmed that the dual mechanism of action of duloxetine is beneficial for the treatment of comorbid conditions, and demonstrated its safety and efficacy in both short- and long-term treatments. Doses up to 120 mg were evaluated across these clinical conditions, demonstrating efficacy and safety in different clinical settings. However, further research on the high-dose range is needed.⁵

Clinical Reasons for Prescribing High-dose Duloxetine

High-dose duloxetine was most commonly prescribed as part of treatment optimization strategies, including dose optimization of a previous lower dose (29.5%), a change in antidepressant (39.5%), or combination with a previous treatment (9.1%), while only a minority of patients received duloxetine as first-line treatment (12.5%). These findings are consistent with recommended therapeutic strategies in MDD, where optimization, switching, and combination approaches are commonly used in patients showing insufficient response to previous treatments. The main reason for prescribing duloxetine in the 90 to 120 mg dose range was the lack of efficacy of previous treatment (41%), followed by a partial response to the 60 mg dose (26.5%) and to the 30 mg dose (4.2%), accounting together for 71.7% of the prescriptions. The existence of tolerability problems with the previous treatment was the reason for prescribing duloxetine in only 5.3% of the patients.

Limitations

This study had a number of limitations. First, the retrospective collection of information implied that the efficacy of duloxetine treatment could not be analyzed on a prospective basis. The collection of aggregated data, in turn, limited the ability to compare intra-subject and inter-subject information. No systematic registry was made of all possible psychiatric and nonpsychiatric comorbidities, so the information may not be complete. It would have been desirable to include a control group for the comparison of data in the same sample of medical consultations of patients treated with duloxetine at doses of 30 to 60 mg, and also patients with other antidepressant treatments. However, this was not considered feasible. The data used to compare our results are derived from references evaluating different populations or nationalities; caution, therefore, should be exercised in interpreting the results, which in any case are of an exploratory nature. The geographical distribution of the 175 participating physicians suggests that generalization of the data is possible, though no sampling was performed for selection.

CONCLUSIONS

In the subjects with MDD treated with high doses of duloxetine (90 to 120 mg), a greater than expected proportion of patients was observed who had advanced, marked, or severe disease, psychiatric comorbidity, sleep disorders, neuropathic pain, and cognitive impairment. The preference for using high doses of duloxetine in psychiatry clinics in these patients allows us to have a better understanding of the profile of those patients who could benefit the most when treating MDD.