Treatment of Depression Associated with Age-related Macular Degeneration in a Double-Blind Randomized Controlled Study

Abstract

BACKGROUND:

Depression is frequently found in patients with age-related macular degeneration (AMD). The purpose of this study was to assess effectiveness of escitalopram in treating major and minor depression in AMD patients.

METHODS:

We conducted a crossover randomized double-blind placebo-controlled 16-week study comparing escitalopram with placebo. Subjects had reduced vision from AMD and also had major or minor depression with a Hamilton Depression Rating Scale (HAM-D₁₇) score ≥10. Subjects were randomized to receive either escitalopram or placebo treatment for 8 weeks and were then crossed over to the other treatment. The primary outcome was change on the total score on HAM-D₁₇ with escitalopram treatment compared to placebo.

RESULTS:

Subjects were 16 AMD patients (mean age 79.1 years) with major (12 subjects) or minor depression (4 subjects), mean HAM-D₁₇ scores = 16.1 (4.2) and mean visual acuity in better eye of 20/70. On escitalopram, subjects showed a significant reduction in HAM-D₁₇ scores compared to placebo treatment (P=.01).

CONCLUSION:

These findings suggest that escitalopram may be an effective treatment option for depressive symptoms associated with major or minor depression in AMD patients with vision loss.

INTRODUCTION

Depression is common in patients with age-related macular degeneration (AMD).^1,2 AMD is the leading cause of permanent vision loss in older adults in the industrial world. It has been described as the epidemic of the baby boomers in the 21^st century.³ Today AMD affects 15-20 million Americans and 40-50 million elderly worldwide. One out of 5 people over the age of 65 has AMD and 1 out 3 over the age of 75. AMD prevention and treatment options are limited to select cases. Patients with AMD are twice as likely to be depressed as elderly community-dwellers who do not have the disease.² Depression contributes more to AMD patients’ disability than their vision loss.^2,4 Results of a small open label study suggested that the selective serotonin reuptake inhibitor (SSRI), sertraline, might be a safe and effective treatment for depression in patients with age-related macular degeneration.⁵

The prevalence of depression is increased in some medical conditions, e.g. stroke, cancer, myocardial infarction, Parkinson’s disease⁶ and congestive heart failure.^7,8 Despite relatively high prevalence rates, depression remains underrecognized and undertreated infarction medical patients.⁹ And yet, depression accounts for poor quality-of-life ratings and has a negative impact on recovery from illness.¹⁰ Further, there is evidence that depression increases mortality risk in certain medical conditions such as coronary heart disease¹¹ and rheumatoid arthritis.¹² There is increasing evidence that antidepressant treatment is as effective in individuals suffering with chronic medical conditions and major depressive disorder as it is in the case of uncomplicated major depression.

The effectiveness of escitalopram has not been studied previously in elderly patients with depression and age-related macular degeneration, nor has there been a randomized double-blind investigation of SSRIs in this population. The primary intent of this study was to determine if depressed AMD patients would show a significant decrease in depressive symptoms with escitalopram treatment compared to placebo in a double-blind randomized controlled trial.

METHODS

Study population and participant flow

We recruited community volunteers from physician offices, the media, an AMD registry, health fairs, senior centers and retirement communities. 18 subjects met criteria below and were enrolled. Two of the 18 enrolled subjects were discontinued by the study psychiatrist within one week of screening for personal reasons, and their data were excluded from analyses. These two subjects did not differ on demographic, macular degeneration or depression rating characteristics from the 16 subjects who completed the study.

The inclusion criteria were diagnosis of age-related macular degeneration; visual acuity 20/40 or worse in the better eye; no other unstable eye disease; depressive disorder (major or minor depression) diagnosed by the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders, Axis I, Fourth (IV) Edition, Research Version–(SCID-IV);¹³ the 17-item Hamilton Rating Scale for Depression (HAM-D₁₇)^14,15 score of ≥10; no cognitive impairment by the Orientation-Memory-Concentration Test (OMC);¹⁶ no hearing problems significant enough to impede assessment; no psychotic disorder; no substance abuse; no history of known allergy to the study drug; no severe or acute medical illness for six months; no history of suicidal or violent behavior; no recent use (2 weeks) of any psychotropic medication; and no initiation of psychotherapy within 3 months prior to study entry.

Design and Procedures

The University of California San Diego Human Research Protections Program approved the protocol for this study (#090013) in December, 2002. Informed consent was obtained prior to screening and enrollment in the study. This was a 16-week AB-BA crossover, randomized, double-blind placebo-controlled study to measure the efficacy of escitalopram in AMD patients with major or minor depression. The study took place in a university department of ophthalmology outpatient setting and was conducted in collaboration with the university’s department of psychiatry. Data for this study came from interviews performed by a trained research associate and study psychiatrist using the measures described below. Masking of the study medications was performed under the direction of a research pharmacist. Study assignment was kept in a locked location and was known only by a research associate not otherwise involved in the study, with provision for the exception in the event of a safety concern. Subjects received the first arm of treatment for 8 weeks and then were directly crossed over. Assessments were performed at baseline, 8 weeks, and 16 weeks.

Baseline assessment

Subjects who met inclusion criteria were enrolled and randomized to receive escitalopram during the first 8-week period followed by placebo during the second 8-week period (the AB group) or to receive placebo during the first period followed by escitalopram during the second period (the BA group). Baseline assessments were performed using the measures below.

8-week assessment and crossover

Subjects were comprehensively evaluated again with the same measures used at baseline. The study psychiatrist assessed the subjects, and subjects were crossed over to the other study treatment.

16-week assessment and exit

Patients were comprehensively evaluated with the same battery that had been used at baseline and at crossover assessments. The study psychiatrist referred patients requiring additional psychiatric care and arranged follow-up dispositions once the protocol was completed.

Monitoring

At weeks 2 and 10, a research associate met with the subjects to monitor compliance. The study psychiatrist monitored response and any adverse effects. At weeks 5 and 13, a research associate telephoned subjects and used a standard questionnaire to monitor and evaluate compliance and adverse effects.

Safety Measures

Standard safety procedures were in place. Subjects were given a 24 hour contact phone number to reach the research associate and study physician in case they had questions about the medication or were experiencing side effects or adverse reactions. All side effects or adverse effects were evaluated and the study psychiatrist was available to provide treatment, make referrals, and/or discontinue study medication, as needed. Subjects were warned about using alcohol or other drugs in combination with the study medication and were warned about the risks of driving. To assure confidentiality of the data, records were kept in double locked files; subjects were assigned project numbers; and no personal identifiers were used in analyzing data or reporting the findings.

Dosing

All subjects began with a 10 mg/day dose of escitalopram/placebo with the option to titrate down to 5mg/day for tolerability.¹⁷

Measures

Primary outcome measure

The primary efficacy outcome measure was the change from start to the end of each 8-week period of this crossover study on the total score of the 17-item Hamilton Rating Scale for Depression (HAM-D₁₇). Scores range from 0-54. The HAM-D₁₇ has been used for four decades as the research standard to assess the severity of depression and response to therapy in clinical investigations. It is based on the clinician’s interview with the patient and probes symptoms such as depressed mood, guilty feelings, suicide, sleep disturbances, anxiety levels and weight loss.

Secondary outcome measures

Secondary efficacy outcome measures included comparative change from start to end of each 8-week period on the HAM-D₁₇ response rates, i.e., >50% reduction in total score, and on the HAM-D₁₇ remission rates, i.e., total score <7.

Other secondary measures were:

Hamilton Rating Scale for Anxiety (HAM-A)¹⁸ total score. The HAM-A is a validated instrument consisting of 14 items scored on a 5-point scale, ranging from 0 (not present) to 4 (severe), to give a total score of between 0 and 56. This widely used scale measures the severity of a patient’s anxiety, based on 14 parameters including mood, tension, fears and somatic complaints. It is useful mainly for assessing a patient’s response to therapy or drug treatment.

The 25-item National Eye Institute Visual Functioning Questionnaire (NEI-VFQ)¹⁹ is widely used to assess impairment in vision-related functioning. Scores range from 1 to 100, with higher scores indicating better functioning. An overall summary scale was created using the average of the 12 subscales.

The Macular Degeneration Self-Efficacy Scale (AMD-SEQ)²⁰ is a 13-item selfreport used to evaluate expectations for handling defined situations related to AMD. Scores range from 1 to 100, with higher scores indicating greater self-efficacy.

Self-rated quality of life²⁰ is a single question using a rating scale of 1 to 100 with higher scores indicating better quality of life.

Measures of clinical and demographic characteristics

Health and Impact Questionnaire.²⁰

Participants were asked about their general health and the impact of macular degeneration on their lives using the Health and Impact Questionnaire. This is a medical history including questions on current medications, living arrangements, and education.

Visual Acuity.

Visual acuity of the better eye and worse eye were obtained using the Snellen chart. Snellen ratings were then converted to the logarithm of the minimum angle of resolution (LogMAR) scale, which is a logarithmic scale on which an increase of 1 point represents a 10-fold drop in vision on the Snellen scale. Whereas 20/20 refers to normal vision and 20/200 to legal blindness on the Snellen scale, using the LogMAR scale, a measurement of 0.0 represents normal vision and 1.0, legal blindness.²¹

Statistical Analysis

Descriptive statistics were used to characterize the sample data at baseline. Continuous variables were compared between the AB and BA groups using two-sample t-tests, or if normality and homogeneity of variance assumptions were not met, Mann-Whitney U tests.

The primary study hypothesis was that escitalopram would be effective in improving depressive symptoms as measured by the primary outcome measure, the total score on the HAM-D₁₇, compared to placebo. Treatment effects for this 2 treatment by 2 period crossover study were tested using a repeated measures ANOVA with sequence (drug then placebo; placebo then drug) as a between-subjects factor and treatment (drug, placebo) and period (baseline, after first treatment, after second treatment) as within-subject factors.²² This statistical method detects effects of the drug as a sequence-period interaction, and any carryover effects as a sequence main effect. Descriptive odds ratios (ORs) were used to compare discrete responses across treatments.

Analysis of treatment effects was performed with the R statistical programming language²³ using the ANOVA function aov. All other descriptive statistics were calculated using Statistica for Windows, version 9.0 (Statsoft, Tulsa, OK). Statistics are reported with 95% confidence intervals, and the significance level for hypothesis testing was set at P < .05.

RESULTS

Patient Characteristics at Baseline

As shown in Table 1, the two groups were similar with regard to demographic and clinical factors at baseline. Mean age of the 16 participants was 79.1 years; 63% were female; mean visual acuity in the best eye was 20/70. Nine of the 16 subjects reported previous use of antidepressants. Eight of these subjects reported that their primary care physician prescribed their antidepressant medication and one subject reported receiving the antidepressant prescription from a mental health professional.

Table 1.

Demographic, socio-economic and clinical characteristics of depressed AMD subjects.

Characteristics	All N= 16	Escitalopram First n= 7	Placebo First n = 9	P-values
Age
Mean (SD) in years	79.1 (4.4)	78.7 (6.6)	79.8 (2.3)	.79
Range	67-86	67-86	76.83
Gender
Female	10 (62.5%)	3 (42.9%)	8 (88.9%)	.11
Male	6 (37.5%)	4 (57.1%)	1 (11.1%)
Years of Education
Mean (SD) in years	15.5 (2.9)	15.6 (2.1)	14.8 (3.7)	.43
Range	12-22	12-22	12-22
Marital Status
Married	7 (43.8%)	2 (28.6%)	4 (44.4%)	.63
Single	9 (56.2%)	5 (71.4%)	5 (55.6%)
Living Arrangement
Alone	8 (50%)	3 (42.9%)	5 (55.6%)	1.00
Living with other(s)	8 (50%)	4 (57.1%)	4 (44.4%)
Visual Acuity
Mean Best LogMAR (SD)	0.54 (0.31)	0.64 (0.42)	0.47 (0.17)	.54
Range	0.30-1.30	0.30-1.30	0.30-0.80
Mean Worst LogMAR (SD)	1.21 (0.67)	1.07 (0.64)	1.31 (0.71)	.52
Range	0.50-2.60	0.50-2.30	0.50-2.60
Hamilton Depression Scale (HAM-D17)				.38
Mean (SD)	16.06 (4.20)	17.10 (4.45)	15.22 (4.02)
Range	10-23	10-23	11-23
Hamilton Anxiety Scale
Mean (SD) (HAM-A)	13.63 (5.20)	14.30 (7.65)	13.11 (2.47)	.69
Range	3-24	3-24	10-18
Visual Function (NEI-VFQ)				.35
Mean (SD) Total Score	57.74 (8.47)	55.64 (7.87)	59.38 (9.01)
Range	44.7-70.2	46.4-66.3	44.6-70.2
Self-Efficacy (AMD-SEQ)
Mean (SD) Total Score	67.96 (17.41)	68.81 (19.95)	67.29 (16.40)	.83
Range	34.4-92.9	39.6-91.5	34.4-92.9
Self-rated quality of life				.30
Mean (SD) Total Score	36.88 (14.82)	31.43 (9.88)	41.11 (17.10)
Range	15.0-60.0	20.0-50.0	15.0-60.0
Number of subjects with prior history of antidepressant use	9/16	4/7	5/9	1.0

Abbreviations: AMD, age-related macular degeneration, HAM-D₁₇, the 17-item Hamilton Rating Scale for Depression; HAM-A, Hamilton Rating Scale for Anxiety; NEI-VFQ, the 25-item National Eye Institute Visual Functioning Questionnaire; AMD-SEQ, the Macular Degeneration Self-Efficacy Scale.

Efficacy Analyses

Primary Outcome

Figure 1 displays individual scores on HAM-D₁₇ at baseline, crossover and exit. (Figure 1). Mean response to escitalopram was 7.6 points greater than response to placebo in the group that received active treatment first, and 6.0 points greater in the group that received placebo first (Table 2). Overall treatment effect of escitalopram was significantly greater than the effect of placebo (sequence by period F statistic = 5.40, P = 0.011). There was a statistically significant sequence effect, suggesting there were carryover effects in the trial (sequence effect F statistic = 4.92, P = 0.032). Net decline in mean HAM-D₁₇ score was 4.7 points in the AB group compared to a net decline of 11.9 in the BA group. By group, subjects receiving active treatment during phase 1 improved during this period and regressed when active treatment was removed, while subjects receiving placebo first improved modestly during period 1, and then further improved when they received active treatment during period 2 (Table 2).

Fig. 1 -.

Individual score at baseline, cross-over, and exit on the Hamilton-D Scale for patients administered Escitalopram 1st (AB) and Placebo 1st (BA) and diagnosed with Major and Minor Depression on SCID-IV

Table 2:

Mean change (SD) in outcome measures by period.

Order	Period 1	Period 2	Difference Across Periods‡	P* value
HAM-D
E 1st: AB†	AB: −6.14 (5.10)	BA: 1.43 (5.62)	−7.57 (10.33)	.01
P 1st: BA††	BA: −2.89 (4.50)	AB: −8.89 (5.62)	6.00 (9.31)
HAM-A
E 1st: AB	AB: −4.00 (7.80)	BA: 3.14 (7.40)	−7.14 (14.80)	.01
P 1st: BA	BA: −3.22 (3.50)	AB: −6.89 (4.00)	3.67 (6.80)
Visual Acuity in LogMAR: Best Eye
E 1st: AB	AB: .07 (.12)	BA: −.06 (.07)	.13 (.19)	.89
P 1st: BA	BA: −.02 (.16)	AB: .01 (.15)	.03 (.31)
Visual Acuity in LogMAR: Worst Eye
E 1st: AB	AB: .08 (.14)	BA: −.06 (.11)	.14 (.19)	.98
P 1st: BA	BA: −.01 (.14)	AB: .01 (.10)	.00 (.19)
NEIVFQ
E 1st: AB	AB: −.83 (4.88)	BA: −2.56 (4.75)	1.73 (6.28)	.83
P 1st: BA	BA: −1.14 (7.65)	AB: 1.11 (4.56)	2.26 (11.00)
Self-Efficacy
E 1st: AB	AB: −2.01 (15.75)	BA: −2.71 (14.46)	0.70 (27.98)	.83
P 1st: BA	BA: 3.97 (11.18)	AB: −4.24 (12.43)	−8.21 (20.90)
Self-rated quality of life
E 1st: AB	AB: 14.29 (13.36)	BA: −9.86 (19.89)	24.14 (25.64)	.12
P 1st: BA	BA: 5.56 (19.28)	AB: 15.00 (17.14)	9.44 (34.04)

^*P-value for treatment effect.

^†E 1^st: AB=Escitalopram first.

^††P 1^st: BA=Placebo first.

^‡For measures where a lower score indicates improvement (e.g., HAM-D₁₇, HAM-A), a negative difference across periods in the AB group and a positive difference across periods in the BA group indications a greater improvement under escitalopram compared to placebo.

Restricting the primary analysis to subjects with major depression had essentially no effect on the analysis, increasing the statistical significance of the overall treatment effect only modestly (p-value decreasing from P = 0.011 when the four subjects with minor depression were included to P = 0.010 when the analysis was restricted to subjects with major depression).

Secondary outcomes

Subjects were more likely to have HAM-D₁₇ scores decline by 50% or more while taking escitalopram than while taking placebo (10 of 16 while on escitalopram versus 2 of 16 while taking placebo, OR = 11.67, 95% CI = (1.94-70.2), P = .009). Patients were more likely to experience remission as measured by a HAM D₁₇ score of <7 points while treated with escitalopram (9 of 16 experiencing remission while on escitalopram versus one of 16 while on placebo, OR = 19.3, 95% CI = (2.03-183.42), P = .006). This difference remained when only patients diagnosed with major depression at baseline were considered (7 of 12 experiencing remission versus 1 of 11, OR = 15.40, 95% CI = (1.50-161.0), P = .03). For the subjects with minor depression, 3 of 4 experienced remission while on escitalopram and 1 of 4 experienced remission while on placebo.

On HAM-A, mean response to escitalopram was 7.1 points greater than response to placebo in the group that received active treatment first, and 3.7 points greater in the group that received placebo first (Table 2). Overall treatment effect of escitalopram was significantly greater than the effect of placebo (sequence by period F statistic = 5.11, P = 0.011). Carryover effects were not evident on the secondary HAM-A outcome (sequence effect F statistic = 2.00, P = 0.16).

In addition, post hoc analyses on self-reported quality of life ratings indicated that patients under escitalopram treatment tended to report more improvement in overall quality of life compared to placebo, P = .12.

There were no serious adverse events related to the protocol. One patient needed to have the dose reduced to 5 mg after 1 week due to insomnia and GI upset; these side effects resolved and thereafter the subject completed the study with no adverse effects. Other minor adverse events included loose stool, falling down a step, transient sweating, and sleepiness, all of which were brief and resolved spontaneously.

DISCUSSION

This 16-week double-blind randomized controlled crossover study of escitalopram and placebo showed that escitalopram had a significant treatment effect on AMD patients with major or minor depression. This was demonstrated by a decrease in severity of depressive symptoms as measured by the HAM-D₁₇ after 8 weeks taking escitalopram compared to 8 weeks taking placebo. Escitalopram treatment compared to placebo also was more likely to result in response, defined as 50% reduction in HAM-D₁₇ scores, and more likely to result in remission.

While consisting of a small number of patients, for the four subjects with minor depression, the improvement in HAM-D₁₇ scores with active drug compared with placebo may be an important finding. Given the increasing evidence for the impact of minor and subsyndromal depressive symptoms on prognosis and course,^24,25 as well as the recent analysis that found that antidepressant treatment effect may correlate with severity of depression,²⁶ our finding that some patients with minor depression may experience significant improvement with antidepressant treatment is noteworthy. There was also a significant treatment effect on anxiety as measured by HAM-A. This finding is consistent with the effects of SSRI treatment.

There were no significant changes on the other secondary measures of function and self-efficacy. A longer course of treatment or additional follow-up may have been needed for these latter secondary measures to show significant change. Alternatively, it is possible that the effects of escitalopram on these latter secondary measures would have been enhanced with concomitant or subsequent treatment with cognitive behavioral therapy, which has also been utilized by our group.^20,27,28

This study was limited to AMD patients with permanently reduced vision who had major or minor depression. We view the depressive symptoms of the study participants as symptoms of depression comorbid with AMD, as a depressive syndrome that has developed concurrently with AMD. People with recurrent depressive disorders are indeed the most vulnerable to relapse and chronicity, especially with the onset of a new medical illness.

Given the AB-BA crossover design, we have considered the possibility that SSRI discontinuation symptoms in the group exposed to escitalopram during period 1 were still present and affected the HAM-D₁₇ scores at the end of period 2. Persistent discontinuation syndrome symptoms can occur with SSRI discontinuation, but were not observed in adverse event monitoring of this trial or reported by subjects, all of whom had a 24 hour telephone contact and were encouraged to report any adverse events or questions.

Another limitation of the crossover design is the possibility of carryover effects or order of treatment effects. Hypothetically our findings could be explained in part by SSRI discontinuation symptoms in the group exposed to escitalopram during period 1 affecting HAM-D measurements at the end of period 2, although, we find this unlikely as discussed above. An alternative explanation for the significant crossover effect observed in our trial is that the improvements in mood during the active treatment period of the AB group washed out when active treatment was removed, while the improvements in mood acquired during the placebo period of the BA group persisted when active treatment was initiated. We note that the best, unbiased estimate of placebo effect is the BA group response to placebo during period 1, where an improvement of 2.9 points was observed in eight weeks. This is less than half the improvement observed under active treatment during period 1 in the AB group (6.1 points in eight weeks) and less than a third of the improvement observed during period 2 in the BA group (8.9 points in eight weeks). Hence, even when we ignore the BA group period 2 data where subjects regressed during placebo treatment, the trial data strongly argue in favor of a positive effect of escitalopram compared to placebo.

Although the results were encouraging, due to the small sample size we cannot definitively conclude that these findings apply to all depressed AMD patients with vision loss in general. Given the evidence for the benefits of effectively treating depression concomitant with other medical illnesses, further research with a larger sample using a randomized placebo-controlled design is warranted to confirm our findings regarding the efficacy of antidepressant treatment and to continue to evaluate for the potential impact on quality of life and functional measures for individuals with impaired vision due to AMD.