Pharmacological treatment for mental health illnesses in adults receiving dialysis: A scoping review

2.1 Study design

We undertook a scoping review that followed the framework of Arksey and O'Malley [17], revised by Levac et al. [18]. Scoping reviews commonly “map” the extent and range of evidence available in the literature with the intent to show the breadth and depth of a topic [18]. Our review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations [19], specifically the extensions for Scoping Reviews (PRISMA-ScR) and Searching (PRISMA-S).

2.2 Search strategy

A systematic literature search, created by an experienced health sciences librarian (MK), was conducted in the following bibliographic databases from inception to February 28, 2022: Medline, EMBASE, PsycINFO via OVID; Cumulative Index for Nursing and Allied Health Literature (CINAHL) via EBSCOhost; Scopus via Elsevier; Web of Science: Core Collection; and Cochrane Library via Wiley. These databases were searched using a combination of natural language vocabulary (keywords) and controlled terms (subject headings). Search terms were derived from three main concepts: (1) broad spectrum of mental health disorders; (2) drug therapy and known pharmaceutical interventions for controlling mental health disorders, including antidepressants, anti-anxiety medications, mood-stabilizing medications, and antipsychotics; (3) CKD, Stage 3 and higher or end-stage, or dialysis. Results were limited to the adult population, and various publication types were removed, including conference materials, books and book chapters, editorials, and letters (see Data S1 for full search strategies by database).

Results from database searches were exported in complete batches on February 28, 2022. Records were imported to the systematic review management software, Covidence. This software was used to identify and remove duplicate records and facilitate title/abstract screening and full-text screening of records.

2.3 Selection criteria

Selection of publications was based on inclusion/exclusion criteria (see Table 1). Texts were required to focus on (a) dialysis modalities, (b) pharmacological treatment including details on dosage and kinetics, and (c) a persistent mental health illness or symptom(s).

2.4 Article selection

We used Covidence to assist 16 reviewers with screening titles, abstracts, and full texts. Reviewers worked in pairs that included at least one pharmacist. Discrepancies among reviewers were discussed for resolution, and the lead authors (JW, PY) were consulted when needed. Team members facilitated translation of the relevant parts of texts written in French, German, and Spanish.

2.5 Data extraction

Following the identification of eligible articles, data were extracted by 15 team members. Pairs consisted of at least one pharmacist; data were extracted by one team member and reviewed by the other. Data extraction for the eligible articles included the following predefined categories: author(s), publication year, patient sex, age, ethnicity, dialysis modality, mental health illness, comorbidities, study design, sample size, study outcome, drug name, dose, frequency, dosing pre- or post-dialysis, route of administration, length of intervention in days, reported adverse effects, and study results. Sample sizes were listed as intention-to-treat.

2.6 Data analysis

A descriptive narrative approach was used to summarize the literature, and only data for patients receiving dialysis were extracted. In Table 2, we present our full findings categorized by DSM-5 criteria for depression, bipolar and related disorders, schizophrenia and psychotic disorders, and anxiety disorders. Table 3 lists antidepressants that were supported by randomized controlled trial (RCT) data. Publications organized by mental health illness and study design are presented in Figure 1.

3.1 Depression

We found 41 articles describing the use of medications for depression in patients receiving dialysis, the majority of whom were receiving hemodialysis (HD) (Table 2). Nineteen publications, ranging from RCTs to case studies, utilized selective serotonin reuptake inhibitors (SSRIs). Sertraline at a dose of 25–200 mg daily was described in nine studies [20, 21, 25, 29, 35, 36, 41, 45, 48]. Mehrotra et al. studied sertraline versus cognitive behavioral therapy (CBT) in a multicenter RCT in patients receiving HD. They noted a statistically significant decrease in the Quick Inventory of Depressive Symptoms-Clinician-Rated (QIDS-C) score at 3 months in the sertraline group, accompanied by a higher incidence of adverse effects [20]. Sertraline was also assessed against placebo in a multicentre, double-blind RCT as an initial feasibility study in HD [21]. No benefit was observed in Beck Depression Inventory-II (BDI-II) and Montgomery–Asberg Depression Rating Scale (MADRS); however, the sample size in this study was small [21]. Two studies assessed sertraline while evaluating the effect of serum cytokines on depression in patients receiving HD [25, 29]. Taraz et al. compared sertraline ranging from 50 to 100 mg versus placebo [25], and Cilan et al. observed the effect of sertraline 50 mg in patients with depression [29]. Although there were inconsistent results in cytokine levels post treatment between the two studies, both authors reported overall improvement in depression scores after initiation of sertraline [25, 29]. Turk and Atalay assessed the effect of sertraline on BDI and QoL measures in patients with depression receiving HD and peritoneal dialysis (PD), respectively [35, 36]. They found sertraline 50 mg daily improved both scores after 8 weeks [36] and 12 weeks [35]. Wuerth et al. conducted a feasibility study including sertraline as a treatment option in patients receiving PD [48]. Seven patients completed 12 weeks of therapy and had an improvement in BDI scores.

Fluoxetine 20 mg daily was assessed in three studies [26, 32, 33]. One placebo-controlled trial noted a statistically significant improvement in BDI, Brief Symptom Inventory (BSI), and electronic visual analog scales at the 4-week midpoint in the treatment group; however, this was not sustained to 8 weeks [26]. Fluoxetine in HD was also assessed against a cohort of patients with normal kidney function, and each group was found to have a reduction in the Hamilton Depression Rating Scale-17 (HAM-D-17) at the end of the 8-week study period [33]. Lee et al explored the impact of fluoxetine 20 mg daily on cytokine levels and nutritional status in 43 patients receiving HD [32]. Of all the cytokines studies, they found only a change in IL-1 beta and IL-6 at the end of the 8-week study period. No change in nutritional status and an overall decrease in Hamilton Depression Rating Scale (HAM-D) was noted in the group. Kauffman reported the use of once weekly fluoxetine ranging in doses from 90 to 180 mg weekly in 4 patients and found this to be effective in managing symptoms of depression [38].

Citalopram 20 mg daily was evaluated in two studies [30, 90]. Hosseini et al. compared citalopram to psychological training to study its effect on depression and anxiety [90]. They did not see a difference in the Hospital Anxiety and Depression Scale (HADS) between the two groups after 3 months. Kalender et al. assessed the effect of citalopram on QoL in patients with CKD/PD/HD with depression and found improvement in two subscales of the Short Form-36 (SF-36) post treatment, also noting a significant decrease in BDI [30]. Escitalopram 20 mg daily plus aerobic exercise was evaluated against either condition alone and then together in one study of 189 patients [23]. After 18 weeks, there was an improvement in QoL, and less patients in the aerobic exercise group were in the severe depression category compared to the escitalopram groups [23].

Paroxetine combined with psychotherapy resulted in an improvement in HAM-D but not on the Zung Self-rating Depression Scale after 8 weeks [31]. Hypoglycemia potentially caused by paroxetine was described in one case report [40]. Fluvoxamine 50 mg nightly was studied in seven patients receiving HD, of which four patients had a 50% decrease in HAM-D scores [44].

Tricyclic antidepressant (TCA) therapy was reported in six publications [27, 37, 49, 52-54], only two of which were trials. In one small placebo controlled trial, amitriptyline was not associated with a statistically significant improvement in the Minnesota Multiphasic Personality Inventory (MMPI), HAM-D, Descriptive Adjective Check List, and physician depressive ratings after 13 weeks [27]. In an observational trial of patients suffering from major depression receiving HD or PD, of seven patients treated with desipramine, four patients recovered from their depressive syndrome, one patient's symptoms improved but didn't resolve, and two patients discontinued due to adverse effects [37]. There were also case reports of the use of amitriptyline [54], amoxapine [49], imipramine [53], nortriptyline [52], and maprotiline [50] although the majority include descriptions of adverse drug reactions [49, 52, 54]. Nefazodone, bupropion, and mirtazapine all demonstrated improvement in depressive symptom measures in the limited number of patients reported [39, 42, 47, 48]. Mianserin and trazodone were discontinued due to their adverse effects [37, 46].

Several other therapies have been studied in patients receiving dialysis with reports on depressive symptoms [22, 24, 28, 34, 51]. Cholecalciferol, compared to placebo, did not significantly improve BDI-II in one RCT in patients receiving HD or PD [22]; Omega-3 fatty acids demonstrated a statistically significant improvement in BDI and SF-36 versus placebo in another RCT in patients receiving HD [24]. S-adenosyl-L-methionine (SAMe) and L-carnitine demonstrated improvement in depression scores compared to baseline in two non-randomized trials in patients receiving HD [28, 34]. Methylphenidate was shown to improve depression scores in two patients receiving dialysis [51].

3.2 Bipolar disorder

We found 16 publications describing the use of medications for bipolar disorder in 17 patients receiving dialysis [60-75]. Nine case reports described lithium therapy with stable oral doses ranging from 300 to 900 mg three times a week post-HD in eight cases [61, 62, 67, 69, 70, 72-74] and 200 mg daily in one case [63]. Two cases described the use of lithium in a non-traditional fashion. Oakley et al. described lithium 0.7–1 mEq/L to be administered in the HD dialysate daily [75], and Flynn et al. described lithium 0.79–0.94 mEq/L administered via continuous ambulatory PD [71]. All cases described improvement of psychiatric symptoms with lithium therapy. Stable lithium serum levels did not exceed 1.5 mmol/L and adverse effects, when reported, included movement abnormalities [70, 75], somnolence with slurred speech [67], vomiting [75], and thirst [43]. Data from case reports described the successful use of valproic acid as part of combination therapy for bipolar disorder in patients receiving HD [60, 61, 65, 68]. Lamotrigine was also described in one case report, resulting in control of hypomanic symptoms [66]. With respect to atypical antipsychotic use for bipolar disorder, three case reports described the use of olanzapine as part of combination therapy in HD [60, 61, 68], and two cases discussed risperidone in HD [64, 68].

3.3 Schizophrenia and psychotic disorders

Eight case reports addressed pharmacological treatments for schizophrenia or schizoaffective disorder in patients receiving dialysis [76-80, 82, 84, 87]. Aripiprazole, as monotherapy or as part of combination therapy, improved psychotic symptoms in two patients receiving HD [76, 78]. Clozapine improved symptoms in two patients with schizophrenia [80, 84]. One patient discontinued clozapine secondary to sedation and initiated paliperidone with clinical improvement noted within three weeks [80]. Risperidone and haloperidol were both reported to improve psychiatric symptoms in patients receiving HD [77, 82, 87]. One case report described occurrences of severe neutropenia with ziprasidone, which may have been dose-related [79]. Four case reports and one review addressed delusional parasitosis therapies in dialysis with the use of risperidone, aripiprazole, olanzapine, and haloperidol [81, 83, 85, 86, 88].

3.4 Anxiety disorders

We found one study that addressed anxiety disorders in patients receiving dialysis. Schohn et al. tested bromazepam in 15 patients receiving HD with anxiety [89]. A final dose range of 3–18 mg per day in divided dosing resulted in an overall improvement of anxiety symptoms. Reported adverse effects included oversedation prior to obtaining the optimal dose.

4.1 Depression

Our review demonstrates that depression was the most extensively studied mental health illness in patients receiving dialysis. However, data describing the treatment effect for pharmacological therapy in this population are heterogeneous. Although most interventional studies report that pharmacological treatments improved depressive symptoms and depression scores, few trials are placebo controlled, and therefore, results are compared to baseline or other therapies (Table 2). The existing studies also utilized a variety of depression tools to assess outcomes, thus making it difficult to compare endpoints across studies.

The following medications have been recommended as first-line treatment in one or more general clinical guidelines for management of depression: SSRIs, serotonin-norepinephrine reuptake inhibitors (SNRIs), bupropion, mirtazapine, vortioxetine, trazodone, vilazodone, agomelatine, and mianserin [92-94]. In alignment with these guideline suggested first-line options, we found that SSRIs have been studied the most robustly in patients receiving dialysis (Table 3). Of the SSRIs, sertraline has been reported at doses ranging from 25 to 200 mg daily [20, 21, 25, 29, 35, 36, 41, 45, 48]; however, the evidence varies substantially from case reports to multicentre trials, with the trial outcomes varying in clinical significance. Other SSRIs that were used in controlled trials to manage depression in patients receiving dialysis include fluoxetine, citalopram, and escitalopram [23, 26, 30]. There was little evidence for the remaining first-line recommended pharmacological therapies for treatment of depression in patients receiving dialysis.

Although there is limited evidence to support the use of antidepressants in patients receiving dialysis [15], the potential use of SSRIs for management of depression in this population has been discussed widely in many reviews [55, 58, 59], with some specifically suggesting sertraline [56, 57]. This could be accomplished by titrating sertraline by 50 mg every two weeks to a maximum of 200 mg daily, with close monitoring of potential adverse effects [57]. This would be followed by a close reassessment of efficacy and discontinuation of therapy if benefit is not achieved.

With the heterogeneity of study types found in our scoping review, there was also a wide variability in the reporting of adverse effects. Mehrotra et al. found that there was a greater incidence of nervous system adverse effects in the sertraline group compared to the CBT group [20]. Taraz et al. found that sertraline carried a higher risk of nausea in patients receiving dialysis than placebo [25]. There were also reports of severe adverse effects possibly attributed to sertraline including serotonin syndrome [41], hospitalization from major bleeding, cardiac, gastrointestinal, or infectious causes [20], cardiac arrest after one dose [21], and death [25]. In a review by Kubanek et al., the use of sertraline in patients receiving dialysis was shown to have a safe impact on QTc in recommended doses, an increased bleeding and fracture risk, a reduction in the risk of intradialytic hypotension, a reduction in uremic pruritus, and a negative impact on sexual function [95]. In a large observational study, Assimon et al also noted differences in risk of sudden cardiac death with the high QT prolonging SSRIs, citalopram and escitalopram, versus fluoxetine, fluvoxamine, paroxetine, and sertraline [96].

Our review found publications for two complementary medicine options that are listed in the Canadian Network for Mood and Anxiety Treatments (CANMAT) guidelines [92]: omega-3 and SAMe [24, 34]. These therapies showed improvement in depressive symptoms in HD [24, 34, 92]; however, further research is needed before these options can be recommended for depression in patients receiving dialysis.

Non-antidepressant treatment options studied for depressive symptoms in patients receiving dialysis included cholecalciferol, methylphenidate, and lithium [22, 43, 51]. The use of cholecalciferol is not concordant with depression management guidelines [92, 93] and was not supported by one RCT in dialysis patients. Clinical guidelines suggest that methylphenidate [92] or lithium [92, 93] may be used as depression adjunctive therapies.

4.2 Bipolar disorder

Publications supporting pharmacological treatments for bipolar disorder in patients receiving dialysis are limited to case reports. The use of lithium is consistent with the joint CANMAT and International Society for Bipolar Disorders (ISBD) guidelines, which recommend its use as first-line treatment for bipolar disorder and acute mania [97]. There are a variety of case reports documenting lithium use with therapeutic drug monitoring in HD [91]. Limited case report data also exist for the use of valproate, lamotrigine, olanzapine, and risperidone in mood stabilization for patients receiving HD [60, 64-66, 68]. These therapies are also supported for use in the CANMAT and ISBD guidelines [97]; however, more data are required.

4.3 Schizophrenia and related disorders

The Canadian Journal of Psychiatry Guidelines for the Pharmacological Therapy of Schizophrenia [98] and American Psychiatric Association Practice Guideline for the Treatment of Patients with Schizophrenia [99] recommend that choice of therapy for schizophrenia be based on a discussion between the patient, caregiver, and health care team, with close assessment of therapy every 2–4 weeks following medication initiation or dose changes. Neither guideline identifies an evidence-based ranking of therapies due to a paucity of evidence informing first-line options, lack of predictability of medication efficacy, and adverse effects. A similar approach could be taken when choosing an appropriate agent in dialysis with recognition of the very limited dialysis data on aripiprazole, risperidone, quetiapine, ziprasidone or haloperidol in treatment-naive patients, and clozapine in treatment-resistant patients.

4.4 Anxiety disorders

Our review found limited evidence to support the use of pharmacotherapeutic options for anxiety in patients receiving dialysis. The 2014 Canadian clinical practice guidelines for the management of anxiety, posttraumatic stress, and obsessive-compulsive disorders list antidepressants (SSRIs, TCAs, SNRIs), benzodiazepines, and buspirone as treatment options [100]; however, more data are required to support their use in patients receiving dialysis.