Cognitive improvement following treatment in late-life depression: relationship to vascular risk and age of onset

Cognitive Improvement Following
Treatment in Late-Life Depression:
Relationship to Vascular Risk and Age
Deanna M. Barch, Ph.D., Gina D’Angelo, Ph.D., Carl Pieper, Dr.P.H,
Consuelo H. Wilkins, M.D., Kathleen Welsh-Bohmer, Ph.D., Warren Taylor, M.D.,
Keith S. Garcia, M.D., Kenneth Gersing, M.D., P. Murali Doraiswamy, M.D.,
Yvette I. Sheline, M.D.
Objectives: To test the hypothesis that the degree of vascular burden and/or age of
onset may influence the degree to which cognition can improve during the course
of treatment in late-life depression. Design: Measurement of cognition both before
and following 12 weeks of treatment with sertraline. Setting: University medical
centers (Washington University and Duke University). Participants: One hundred
sixty-six individuals with late-life depression. Intervention: Sertraline treatment.
Measurements: The cognitive tasks were grouped into five domains (language, pro-
cessing speed, working memory, episodic memory, and executive function). We mea-
sured vascular risk using the Framingham Stroke Risk Profile measure. We mea-
sured T2-based white matter hyperintensities using the Fazekas criteria. Results:
Both episodic memory and executive function demonstrated significant improvement
among adults with late-life depression during treatment with sertraline. Importantly,
older age, higher vascular risk scores, and lower baseline Mini-Mental State Exami-
nation scores predicted less change in working memory. Furthermore, older age, later
age of onset, and higher vascular risk scores predicted less change in executive func-
tion. Conclusions: These results have important clinical implications in that they
suggest that a regular assessment of vascular risk in older adults with depression is
necessary as a component of treatment planning and in predicting prognosis, both
for the course of the depression itself and for the cognitive impairments that often
accompany depression in later life. (Am J Geriatr Psychiatry 2012; 20:682–690)
Key Words: Cognition, treatment, vascular depression, white matter
Received November 11, 2010; revised February 25, 2011; accepted April 08, 2011. From the Departments of Psychology (DMB), Psychiatry(DMB, CHW, KSG, YJS), Radiology (DMB, YJS), and Division of Biostatistics (GD’A), Washington University School of Medicine, St. Louis,MO; and Department of Psychiatry, Duke University School of Medicine, Durham, NC (CP, KW-B, WT, KG, PMD). Send correspondenceand reprint requests to Deanna M. Barch, Ph.D., Departments of Psychology, Psychiatry, and Radiology, Washington University, Box 1125,One Brookings Dr, St. Louis, MO 63130. e-mail: [email protected] Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.AJGPonline.org).
C 2012 American Association for Geriatric Psychiatry Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this Numerous studies document the presence of Such findings in younger adults raise the possi- a range of cognitive impairments in late-life bility that at least some aspects of cognitive dys- depression,1,2 including reductions in working mem- function in late-life depression may not necessarily ory, executive function, episodic memory, and pro- result from accruing vascular changes but may reflect cessing speed.1,2 One hypothesis is that the presence state or even trait aspects of depression. If so, one of such cognitive impairments in late-life depression might expect more robust evidence for improvement reflects frontal striatal and/or hippocampal dysfunc- in cognitive function across the course of treatment tion that may result—at least in part—from vascu- in late-life depression. Several studies have found lar disease,3–6 and that this may be particularly true some evidence for such improvement, with the mag- for depression with a later age of onset. Given this nitude of improvement in cognitive function corre- hypothesis, it is not surprising that cognitive impair- lated with the magnitude of improvement in depres- ment persists following depression treatment in older sion in some studies.19–25 However, a number of other adults (for a review, see Douglas and Porter7). As such, studies have either found no improvement in cogni- the goal of this article was to examine the influence of tion as a function of treatment in late-life depression vascular risk and age of onset on cognitive improve- or that the level of cognitive function in antidepres- ment following depression treatment in older adults.
sant responders was still below that of individuals that more severe cognitive impairment among older One important factor that these studies have not adults with depression predicts poorer outcome8,9 taken into account is the role that vascular burden and a poorer response to treatment.10–13 Furthermore, and the presence of white matter hyperinstensities older adults with greater evidence of white matter may play in moderating cognitive change. Some stud- impairment13,14 or reduced hippocampal and cor- ies have found that older adults with depression are tical volumes15,16 also show a poorer response to either less likely to respond to antidepressant treat- treatment. All of these results are consistent with ment or slower to respond.31 This could reflect the the hypotheses that cognitive impairments in late- influence of vascular changes and white matter alter- life depression result from vascular or other neural ations in older adults with depression. If so, then the changes that contribute to the onset or recurrence of degree vascular burden, white matter hyperintensi- depression in late life and are not transient or state- ties, and/or age of onset may influence the degree related manifestations of the presence of depression. If to which cognition can improve during treatment in so, it is not surprising that such cognitive deficits per- late-life depression. The goal of this study was to test sist even among individuals who respond to depres- these hypotheses by examining the degree to which vascular risk, white matter hyperintensities, and age However, there is also evidence that cognitive of onset predict the degree of cognitive improvement impairments are present in depression with an ini- during 12 weeks of sertraline treatment in a large sam- tial onset in young adulthood or middle age. These ple of older adults with Diagnostic and Statistical Man- deficits are often present in many of the same cog- ual of Mental Disorders, Fourth Edition (DSM-IV), Major nitive domains impaired in late-life depression,17 though the robustness of such cognitive impair-ments in early or midlife adult depression has beenmixed across studies.18 Furthermore, there is evi- dence that at least some of these cognitive impair- Participants
ments can improve during treatment in youngerdepressed adults. Specifically, Douglas and Porter7 Participants were recruited as part of a National recently reviewed this literature and concluded that Institute of Mental Health–funded study through measures of episodic memory, verbal fluency, and advertising and physician referral to Washington Uni- processing speed varied as a function of clinical state versity (WU) and Duke University (Duke). Patients in depression, with deficits in executive function and were recruited into the study if they met DSM-IV cri- attention showing more stable and trait-like charac- teria for MDD by Structured Clinical Interview for teristics in younger adults with depression.
Axis I DSM-IV Disorders (SCID-IV),32 administered Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this Cognitive Change in Late-Life Depression by a research psychiatrist, and were 60 years or older.
(MADRS).35 The MADRS was administered by a Exclusion criteria included 1) severe or unstable medi- research psychiatrist at the start of the trial and cal disorders; 2) known primary neurologic disorders; during each week of the trial. We assessed overall 3) history of other Axis I disorders prior to the depres- cardiovascular risk using the Framingham Stroke sion diagnosed by SCID-IV; 4) current suicidal risk; 5) Risk Profile.36 The Framingham Stroke Risk Profile a current MDD episode that had failed to respond to generates a composite score using the following adequate trials of two prior antidepressants for at least vascular risk factors to predict 10-year probability 6 weeks at therapeutic doses; 6) use of psychotropic of stroke in both men and women: age, systolic prescription or nonprescription drugs or herbals (e.g., blood pressure, the use of antihypertensive therapy, Hypericum) within 3 weeks or 5 half-lives; 7) inpatient diabetes mellitus, cigarette smoking, cardiovascular status; or 8) Clinical Dementia Rating greater than 0 or disease (coronary heart disease, cardiac failure, or Mini-Mental State Examination score less than 21.27 Of intermittent claudication), atrial fibrillation, and 362 phone screens at WU and 374 at Duke, there were left ventricular hypertrophy by electrocardiogram.
181 clinic screens at WU and 135 at Duke (for details, This score has been positively associated with white see Sheline et al.13). The 316 clinic screens resulted in matter hyperintensities37 and negatively associated 217 participants (120 at WU and 97 at Duke) being with total brain volume.38 We also assessed baseline enrolled in a 12-week treatment trial with sertraline.
global cognitive function using the Clinical Dementia Of these participants, 190 completed treatment (109 Rating39 and Mini-Mental State Examination.40 We at WU and 81 at Duke). Written informed consent assessed age at onset from the SCID-IV and all approved by the relevant institutional review board available medical and psychiatric records.
was obtained for all subjects. This trial is registeredat clinical trials.gov Treatment Outcome of Vascular Neuropsychological Function
Participants were administered a large battery Sertraline Treatment
of neuropsychological tests that covered cognitivedomains relevant to understanding late-life depres- Sertraline was selected as the SSRI in this sion at both baseline (prior to the start of medications) study because it is among the more selective and at the end of the 12 weeks of treatment. The neu- 5-hydroxytryptamine (5-HT; also serotonin) re- ropsychological testing was performed by a trained uptake inhibitors, has an excellent profile for safety examiner who was supervised by a Ph.D.-level psy- and effectiveness in the treatment of MDD in the con- chologist (DB and KWB). We grouped the cognitive text of comorbid illness,33 has linear kinetics, and has tasks into rationally motivated domains described minimal age effect on clearance.34 The primary results later, based on the prior literature regarding the cog- of depression response to treatment have been previ- nitive processes tapped by each of the tasks. The ously reported.13 Briefly, the treatment consisted of domains were executive function, processing speed, an initial dose of sertraline at 25 mg for 1 day to rule episodic memory, working memory, and language out drug sensitivity and then 50 mg daily, with sub- processing. To combine the tasks within each cog- sequent dose changes at 2, 4, and 6 weeks (to 100, nitive domain, we created Z-scores for the primary 150, and 200 mg per day, respectively). At any point, dependent measure of interest using the scores from patients who had side effects could be titrated to a both baseline and follow-up across all participants lower dose. Medication adherence was assessed on and then summed the Z-scores (the results would not each visit by self-report. At the end of treatment, the have been different if only the baseline was used to mean final dose was 114, with 64 on less than 100 mg, create Z-scores). For the majority of variables, a higher 60 on 100 to 125 mg, 46 on 150 to 175 mg, and 34 on score indicated better performance. We reverse scored any items (e.g., reaction time on Trails B) for which Measures
good performance was indicated by a lower value.
Cronbach’s α (a measure of internal consistency) was We assessed depression severity using the computed for each domain from the baseline data.
For further details, see Sheline et al.,2 who describe Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this the baseline analyses of neuropsychological function School of Medicine by RCM and YIS. The modi- fied Fazekas criteria35 describe magnetic resonance Executive function. This domain included verbal imaging hyperintensities in three regions (periven- fluency (total phonological and semantic), Trails B tricular, deep white matter, and subcortical gray mat- (reverse scored time to completion), the color-word ter regions), using ascending degree of severity. The interference condition of the Stroop task (number dependent variable was a total score summing sever- completed), the Initiation-Perseveration subscales of ity scores in all three regions. See Supplemental Digi- the Mattis, and categories completed from the Wis- tal Content 1 (http://links.lww.com/AJGP/A28) for consin Card Sorting Test. The coefficient α for this Processing speed. This domain included symbol- digit modality (number completed), the color nam-ing condition of the Stroop task (number completed), and Trails A (reverse scored time to completion). Thecoefficient α for this domain was 0.80.
Of the 217 individuals enrolled in the trial, 211 had Episodic memory. This domain included word list usable neuropsychological data at baseline. Of the 190 learning (total correct), logical memory (total correct individuals who completed the trial, 166 (105/109 at immediate), constructional praxis (Rey-Osterrieth WU, 61/81 at Duke) were able to provide cognitive Complex Figure Test memory performance), and the data at the 12-week follow-up. We began by com- Benton Visual Retention Test (total correct). The coef- paring the demographic, clinical, and cognitive char- ficient α for this domain was 0.76.
acteristics of the participants with usable neuropsy- Language processing. This domain included the chological data at baseline (n = 211) who did (n = Shipley Vocabulary Test (number correct), the Boston 166) and did not have data at study completion.13 As Naming Test (number correct), and the word reading shown in Table 1, these two groups did not differ in condition of the Stroop task (number completed). The age, education, gender, race, baseline MADRS, base- coefficient α for this domain was 0.67.
line Mini-Mental State Examination scores, vascular Short-term/working memory. This domain included risk, or on any of the 5 cognitive domains. The indi- digit span forward (number of trials correctly com- viduals who did not have neuropsychological data at pleted), digit span backward (number of trials cor- study completion had a slightly but significantly later rectly completed), and ascending digits (number of age of onset than those individuals who did not.
trials correctly completed). The coefficient α for thisdomain was 0.68.
Did Cognition Improve Across the Course
Magnetic Resonance Imaging
of Treatment?
Both T1 and T2 magnetic resonance images were To examine whether performance in any of the collected using a Siemens Sonata 1.5-T scanner at five cognitive domains improved across treatment, WU School of Medicine and a GE 1.5-T scan- we used a repeated-measures analysis of variance ner at Duke. See Supplemental Digital Content 1 with time point (baseline, follow-up) and cognitive (http://links.lww.com/AJGP/A28) for details on domain (language, processing speed, working mem- ory, episodic memory, executive function) as within-subject factors, and the domain Z-score as the depen- T2-Weighted Hyperintensities
dent measure. This analysis revealed a main effect oftime, F(1,165) = 19.3, p <0.0001, and a time by cog- Hyperintensities were assessed blinded to treat- nitive domain interaction, F(4,660) = 11.1, p <0.001.
ment data using the modified Fazekas criteria, which As shown in Figure 1, follow-up contrasts for each are widely used measures of white matter burden cognitive domain indicated that only episodic mem- that allows comparison with a large number of pre- ory and executive function improved over the course vious studies. All ratings were conducted at WU Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this Cognitive Change in Late-Life Depression TABLE 1. Demographic, Clinical, and Cognitive Characteristics of Participants
Mean (SD)
Comparison of
Completers and
Variable
Completers
Noncompleters
Noncompleters
t(209) = 2.06, p = 0.04
Notes: bold value indicates the only variable that differed between completers and non-completers.
MMSE, Mini-Mental State Examination.
thus we used Spearman rank order correlations to FIGURE 1. Graph of performance in each of the five
cognitive domains at baseline and at
examine the relationship with cognitive change for posttreatment. Significance of change in each
this variable. As shown in Table 2, older age pre- cognitive domain was assessed with post-hoc
dicted less improvement in processing speed, work- contrast (F-tests with 1,165 dfs).
ing memory, and executive function. Later age ofonset predicted less improvement in executive func-tion. Higher vascular risk predicted less change inworking memory and executive function. More severewhite matter hyperintensities predicted less change inprocessing speed. Of note, the results were essentiallyidentical when partial correlations were used, exam-ining the relationship between posttreatment perfor-mance and the predictors, covarying for baseline cog-nitive performance.
We also examined whether the magnitude of improvement in depression predicted the magnitudeof improvement in cognition. To do so, we created a residualized change score for depression usingbaseline MADRS scores to predict posttreatment What Factors Predict Improved Cognition?
MADRS scores and correlated this with the residu-alized change scores for cognition. As shown in Table We computed residualized change scores for each 2, a greater reduction in MADRS scores predicted a of the five cognitive domains, using baseline perfor- greater improvement in language function but did not mance to predict posttreatment performance. We then predict improvement in the other cognitive domains.
used Pearson’s product moment correlations to com-pute the correlations between age, age of onset, vascu- Cognitive Improvement As a Function of
lar risk score, baseline Mini-Mental State Examination Remitter Status
scores, and baseline MADRS depression scores withthe 5 cognitive domain residualized change scores.
Some prior research suggests that cognitive Fazekas scores were not normally distributed and improvement over the course of treatment may vary Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this TABLE 2. Correlations of Predictor Variables With Residualized Change Scores (Baseline to Posttreatment) in Each
Cognitive Domain
Variable
Language Function
Processing Speed
Working Memory
Episodic Memory
Executive Function
− 0.18a
− 0.25c
− 0.17a
− 0.24c
− 0.24c
− 0.17a
− 0.16a
− 0.16a
− 0.19a
Note: N = 166. Bold variables indicate siginificance and p>.05 or less. MMSE, Mini-Mental State Examination.
as a function of whether an individual was considered older individuals with depression and to determine to have remitted to treatment in terms of depression whether factors such as the degree of vascular burden, status.22,30 Thus, we examined cognitive change as a white matter hyperintensities, and/or age of onset function of remitter status, with a remitter defined as influenced the degree to which cognition improved someone who achieved a final MADRS score of 7 or during treatment in late-life depression. We found less. Of the 166 individuals with neuropsychological that both episodic memory and executive function data at both baseline and follow-up, 63 were remitters improved from baseline to posttreatment and that this and 103 were not. We computed a repeated-measures improvement occurred for individuals whose depres- analysis of variance with time (baseline, posttreat- sion remitted and for those whose depression did ment) and cognitive domain as within-subject factors not remit. However, working memory improved only and remitter status as a between-subject factor. This among individuals with depression whose depres- analysis of variance again revealed a time by cog- sion remitted. Of note, we cannot definitely attribute nitive domain interaction, F(4,656) = 9.92, p <0.001 these changes to the treatment, as we did not have and revealed a main effect of remitter status, F(1,164) a placebo control group. However, importantly, we = 8.37, p = 0.004. However, there was no signifi- found that a number of factors moderated the degree cant two-way interaction between remitter status and of improvement in cognition (whether it was specif- time, F(1,164) = 1.24, p = 0.27, or three-way interaction ically due to treatment or responsivity to practice) between remitter status, time, and cognitive domain, particularly among those individuals whose depres- F(4,656) = 1.56, p = 0.184. As shown in Figure 2, the sion did not fully remit. Specifically, older age, higher main effect of remitter status reflected the fact that vascular risk scores, and lower baseline Mini-Mental the nonremitters had overall worse cognitive perfor- State Examination scores predicted less improvement mance than the remitters at both baseline and post- in working memory. Furthermore, older age, later age of onset, and higher vascular risk scores predictedless improvement in executive function. In addition,more severe white matter hyperintensities predicted DISCUSSION
less improvement in processing speed.
The fact that episodic memory improved is consis- The goal of the current study was to examine the tent with prior work, suggesting that impairments degree to which cognitive function improved dur- in episodic memory may be associated with state ing the course of antidepressant treatment among components of depression and may be more likely Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this Cognitive Change in Late-Life Depression FIGURE 2. Graph of performance in each of the five cognitive domains at baseline and at posttreatment, plotted separately for
those individuals whose depression remitted by the end of treatment (MADRS score ≤7) and for those individuals
who depression did not remit by the end of treatment (see Supplemental Digital Content 1,
http://links.lww.com/AJGP/A28).
to improve with the remission of depression than sistent with the hypothesis that vascular changes lead- some other cognitive functions.7 However, we also ing to white matter alternations may contribute to saw improvements in executive function, a domain at least some of the cognitive impairments found in in which improvements have not been as consistently late-life depression.3–6 It is somewhat surprising that demonstrated in prior studies.7 This is a cognitive vascular burden and not white matter hyperinten- domain that Douglas and Porter argued may reflect sities predicted change in some of the other cogni- more trait-like aspects of depression and which has tive domains such as executive function and work- been associated with white matter abnormalities in ing memory. However, it may be that our measure of late-life depression.3 Interestingly, the predictors of white matter hyperintensities, which was restricted change in episodic memory and executive function to periventricular, deep white matter, and subcortical were very different. The degree of change in execu- gray matter regions, did not capture changes in white tive function, but not episodic memory, was predicted matter in other brain regions that may also be related by older age, older age of onset, and higher vascular risk scores. In contrast, baseline depression severity We also found that individuals whose depression predicted change in episodic memory but not exec- remitted during treatment showed overall better cog- utive function. Similar to executive function, change nitive function. This result is consistent with our prior in working memory was also predicted by older age work (in this same sample) showing that baseline and higher vascular risk scores. Thus, although both cognitive function predicted response to treatment13 episodic memory and executive function improved and with other work showing that impaired cogni- over the course of treatment, the predictors of the tive function in late-life depression is associated with magnitude of change in executive function are con- sistent with a critical role for vascular burden in There were several limitations to this study. First, constraining executive function (as well as working we did not recruit a control sample of older adults memory) and the degree to which it can improve in without depression, as the purpose of the study was to examine treatment response in older adults In our prior work with this sample, we found that with depression. Thus, we could not directly address more severe white matter hyperintensities were asso- the question of whether cognitive function in our ciated with worse cognitive function in all domains depressed individuals was worse than controls at at baseline.13 Interestingly, in the current analyses, baseline or whether the degree of cognitive improve- we also found that more severe white matter hyper- ment would have resulted in a level of cognitive intensities predicted less improvement in processing performance that no longer differed from individ- speed. This relationship with processing speed is con- uals without depression. However, the large body Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this of research showing that late-life depression is asso- ment that can be obtained in some domains among ciated with impaired cognition relative to matched older adults treated for depression. These results have controls makes it likely that we would have also important clinical implications in that they suggest found that our sample was impaired relative to an that a regular assessment of vascular risk in older appropriate control group. Second, we did not have adults with depression is necessary as a component a placebo control group, so we cannot definitely of treatment planning and in predicting prognosis, determine that the cognitive change we did observe both for the course of the depression itself and for the reflected a response to treatment rather than prac- cognitive impairments that often accompany depres- tice effects or placebo effects. However, this does not minimize the importance or utility of identify-ing predictors of cognitive change, regardless of the The authors thank Dan Blazer M.D., Ph.D., for serv- source of change. In other words, the ability to ben- ing as an advisor to the study, Caroline Hellegers, M.A., efit from practice may be key to various cognitive for her assistance with study coordination at Duke Univer- enhancement approaches and thus information about sity and Tony Durbin, M.S., and Brigitte Mittler for their the factors that may identify who or who would assistance with study coordination at Washington Univer- show responsivity to either antidepressant therapy sity. Drs. Sheline, Doraiswamy, and Taylor have received or practice may be useful in individualized treatment grants and/or speaking/consulting fees from antidepres- sant manufacturers but do not own stock in these compa- In summary, we found that older adults with nies. Dr. Krishnan is also a coinventor on a patent that is depression showed significant improvement in licensed to Cypress Biosciences and owns stock in CeneRx. episodic memory and executive function across the Dr. Doraiswamy also owns stock in EnergyInside. course of a 12-week treatment with sertraline. Further- This work was supported by a Collaborative R01 for more, we found that factors such as age, age of onset, Clinical Studies of Mental Disorders Grant MH60697 and vascular risk scores predicted the amount of (YIS) and MH62158 (PMD). YIS also received support change in cognitive domains such as executive func- from NIMH K24 65421. In addition, this work was sup- tion and working memory, a result consistent with the ported by a grant (RR00036) to the WUSM General Clin- hypothesis that vascular burden may play a critical ical Research Center and by a grant from Pfizer, Inc., to role in constraining the degree of cognitive improve- References
1. Butters MA, Whyte EM, Nebes RD, et al: The nature and deter- 8. Baldwin RC, Gallagley A, Gourlay M, et al: Prognosis of late life minants of neuropsychological functioning in late-life depression.
depression: a three-year cohort study of outcome and potential Arch Gen Psychiatry 2004; 61(6):587–595 predictors. Int J Geriatr Psychiatry 2006; 21(1):57–63 2. Sheline YI, Barch DM, Garcia K, et al: Cognitive function in late life 9. Potter GG, Kittinger JD, Wagner HR, et al: Prefrontal neuropsycho- depression: relationships to depression severity, cerebrovascular logical predictors of treatment remission in late-life depression.
risk factors and processing speed. Biol Psychiatry 2006; 60(1):58– Neuropsychopharmacology 2004; 29(12):2266–2271 10. Alexopoulos GS, Kiosses DN, Heo M, et al: Executive dysfunc- 3. Sheline YI, Price JL, Vaishnavi SN, et al: Regional white mat- tion and the course of geriatric depression. Biol Psychiatry 2005; ter hyperintensity burden in automated segmentation distin- guishes late-life depressed subjects from comparison subjects 11. Sneed JR, Keilp JG, Brickman AM, et al: The specificity of neu- matched for vascular risk factors. Am J Psychiatry 2008; 165(4): ropsychological impairment in predicting antidepressant non- response in the very old depressed. Int J Geriatr Psychiatry 2008; 4. Alexopoulos GS: Vascular disease, depression, and dementia. J Am 12. Story TJ, Potter GG, Attix DK, et al: Neurocognitive correlates of 5. Alexopoulos GS: The vascular depression hypothesis: 10 years response to treatment in late-life depression. Am J Geriatr Psychi- later. Biol Psychiatry 2006; 60(12):1304–1305 6. Barnes DE, Alexopoulos GS, Lopez OL, et al: Depressive symp- 13. Sheline YI, Pieper CF, Barch DM, et al: Support for the vascular toms, vascular disease, and mild cognitive impairment: findings depression hypothesis in late-life depression: results of a 2-site, from the Cardiovascular Health Study. Arch Gen Psychiatry 2006; prospective, antidepressant treatment trial. Arch Gen Psychiatry 7. Douglas KM, Porter RJ: Longitudinal assessment of neuropsycho- 14. Alexopoulos GS, Kiosses DN, Choi SJ, et al: Frontal white matter logical function in major depression. Aust N Z J Psychiatry 2009; microstructure and treatment response of late-life depression: a preliminary study. Am J Psychiatry 2002; 159(11):1929–1932 Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this Cognitive Change in Late-Life Depression 15. Simpson SW, Baldwin RC, Burns A, et al: Regional cerebral vol- 27. Nebes RD, Butters MA, Mulsant BH, et al: Decreased working ume measurements in late-life depression: relationship to clinical memory and processing speed mediate cognitive impairment in correlates, neuropsychological impairment and response to treat- geriatric depression. Psychol Med 2000; 30(3):679–691 ment. Int J Geriatr Psychiatry 2001; 16(5):469–476 28. Bhalla RK, Butters MA, Mulsant BH, et al: Persistence of neuropsy- 16. O’Brien JT, Lloyd A, McKeith I, et al: A longitudinal study chologic deficits in the remitted state of late-life depression. Am J of hippocampal volume, cortisol levels, and cognition in older depressed subjects. Am J Psychiatry 2004; 161(11): 29. Nakano Y, Baba H, Maeshima H, et al: Executive dysfunction in medicated, remitted state of major depression. J Affect Disord 17. Castaneda AE, Tuulio-Henriksson A, Marttunen M, et al: A review on cognitive impairments in depressive and anxiety disorders with 30. Culang ME, Sneed JR, Keilp JG, et al: Change in cognitive function- a focus on young adults. J Affect Disord 2008; 106(1–2):1–27 ing following acute antidepressant treatment in late-life depres- 18. Grant MM, Thase ME, Sweeney JA: Cognitive disturbance in outpa- sion. Am J Geriatr Psychiatry 2009; 17(10):881–888 tient depressed younger adults: evidence of modest impairment.
31. Zanardi R, Cusin C, Rossini D, et al: Comparison of response to fluvoxamine in nondemented elderly compared to younger 19. Savaskan E, Muller SE, Bohringer A, et al: Antidepressive ther- patients affected by major depression. J Clin Psychopharmacol apy with escitalopram improves mood, cognitive symptoms, and identity memory for angry faces in elderly depressed patients. Int 32. First MB, Spitzer RL, Gibbon M, et al: Structured clinical interview J Neuropsychopharmacol 2008; 11(3):381–388 for the DSM-IV-TR Axis I disorders. Washington, DC, American 20. Gallassi R, Di Sarro R, Morreale A, et al: Memory impairment in patients with late-onset major depression: the effect of antidepres- 33. Cipriani A, La Ferla T, Furukawa TA, et al: Sertraline versus other sant therapy. J Affect Disord 2006; 91(2–3):243–250 antidepressive agents for depression. Cochrane Database Syst Rev 21. Doraiswamy PM, Krishnan KR, Oxman T, et al: Does antidepres- sant therapy improve cognition in elderly depressed patients? J 34. DeVane CL, Pollock BG: Pharmacokinetic considerations of Gerontol A Biol Sci Med Sci 2003; 58(12):M1137–M1144 antidepressant use in the elderly. J Clin Psychiatry 1999; 60(suppl 22. Devanand DP, Pelton GH, Marston K, et al: Sertraline treatment of elderly patients with depression and cognitive impairment. Int 35. Montgomery SA, Asberg M: A new depression scale designed to J Geriatr Psychiatry 2003; 18(2):123–130 be sensitive to changes. Br J Psychiatry 1979; 134:382–389 23. Butters MA, Becker JT, Nebes RD, et al: Changes in cognitive func- 36. Wolf PA, D’Agostino RB, Belanger AJ, et al: Probability of stroke: a tioning following treatment of late-life depression. Am J Psychiatry risk profile from the Framingham Study. Stroke 1991; 22:312–318 37. Jeerakathil T, Wolf PA, Beiser A, et al: Stroke risk profile pre- 24. Beats BC, Sahakian BJ, Levy R: Cognitive performance in tests sen- dicts white matter hyperintensity volume: the Framingham Study.
sitive to frontal lobe dysfunction in the elderly depressed. Psychol 38. Seshadri S, Wolf PA, Beiser A, et al: Stroke risk profile, brain vol- 25. Abas MA, Sahakian BJ, Levy R. Neuropsychological deficits and ume, and cognitive function: the Framingham Offspring Study.
CT scan changes in elderly depressives. Psychol Med 1990; 39. Morris JC: The Clinical Demential Rating (CDR): current version 26. Nebes RD, Pollock BG, Houck PR, et al: Persistence of cognitive and scoring rules. Neurology 1993; 43:2412–2414 impairment in geriatric patients following antidepressant treat- 40. Folstein MF, Folstein SE, McHugh PR: Mini-Mental State: a practical ment: a randomized, double-blind clinical trial with nortriptyline method for grading the cognitive state of patients for the clinician.
and paroxetine. J Psychiatr Res 2003; 37(2):99–108 Am J Geriatr Psychiatry 20:8, August 2012 Copyright American Association for Geriatric Psychiatry. Unauthorized reproduction of this

Source: http://ccpweb.wustl.edu/pdfs/Barch2012CognitiveImprovementFollowingTreatment.pdf