Zheli CFEN*， Guanghua LAN， Xinhua SFEN， Xingen PAN， Xiaoxun CFEN， Jianhua LI

·Original article·

Relationship of changes in cognitive and depressive symptoms during anti depressant treatment of individuals with geriatric depression and their relationship to the APOE epsilon 4 allele

Zheli CFEN1*， Guanghua LAN2， Xinhua SFEN1， Xingen PAN1， Xiaoxun CFEN1， Jianhua LI1

1. Introducti on

Cogniti ve impairment is common in geriatric depression.[1，2]Some authors suggest that the cognitive impairment in geriatric depression does not improve with remission of depression， and it may even progress to fully-blown dementi a.[3]One rationale for such a relati onship is that Alzheimer’s Disease and late-onset depression share a common genetic risk factor， the epsilon 4 allele of the apolipoprotein E gene （APOE）.[4，5]It remains unclear if the cognitive impairment seen in some geriatric patients with depression is independent of depressive symptoms， one of the symptoms of the depressive episode， or both. We hypothesize that the cognitive impairment seen in pati ents with geriatric depression persists after depressive symptoms resolve， particularly in those with the epsilon 4 allele of APOE. To assess this hypothesis we classif i ed elderly individuals with a first episode of depression into those with and withoutco-occurring mild cognitive impairment （NCI） and then followed the one-year trajectory of cogniti ve and depressive symptoms in these pati ents as they received anti depressant treatment for their depression.

2. Methods

2.1 Sample identification

The fl owchart for the study is shown in Figure 1. This is a prospective study of patients with first-onset geriatric depression who sought outpatient or inpatient treatment at the Third People’s Fospital of Fuzhou from January 2009 to December 2010. Included patients were at least 60 years of age， met diagnostic criteria for depression specif i ed in the 3rdedition of the Chinese Classification of Mental Disorders （CCND-3），[6]had a score of at least 11 on the Geriatric Depression Scale （GDS），[7]had had no prior episode of depression，had no history of alcohol or drug dependence， had no serious medical condition （including neurodegenerative disorders， cerebrovascular disease， or serious diseases of the heart， liver or kidney） and signed a consent form to participate in the study. The 64 enrolled patients who completed the 12-month follow-up included 17 males and 47 females with a mean age of 67.5 （s.d.=6.0） years and a mean duration of education of 8.9 （4.2） years.

The criteria proposed by Petersen[8]were used to identi fy depressed patients with NCI: （a） impaired memory reported by the pati ent or their relati ves；（b） a memory test score of at least 1.5 standard deviations lower than normal controls with the same age and educati onal level； （c） a grade of 2 to 3 in the Global Deterioration Scale （GDS）[9]； （d） normal general cognitive functioning and daily functi oning with an Acti vity of Daily Living （ADL）[8]score lower than 26；and （e） no diagnosis of dementia or any other physical or mental illness that may lead to brain dysfuncti on.Among the 64 pati ents with geriatric depression who completed the 12-month follow-up， 30 （46.9%） met these criteria for NCI.

Control subjects were recruited from pati ents 60 years or older seeking a routine physical examinati on at the Geriatric Department or the Physical Examinati on Center of the Third People’s Fospital of Fuzhou.Individuals with a serious physical illness， those who met the above criteria for depression or NCI and those with a family history of depression， dementia or NCI in first-degree relatives were excluded. The 31 recruited controls， all of whom signed consent forms，included 13 males and 18 females， had a mean age of 68.2 （8.6） years and a mean durati on of educati on of9.0 （4.0） years. There were no statistically significant differences between patients and controls in age （t=-0.415， p=0.687）， gender （χ2=2.28， p=0.131）， or years of education （t=-1.09， p=0.280）.

Figue 1. Flowchart for the study

2.2 Assessment and treatment

Study participants were administered the GDS， the Nini-Nental State Exam （NNSE），[10]the forward and backward digit span task， and the Trail-Naking Tests A and B （TNT-A， TNT-B）[11]at baseline and the pati ent group was re-administered these tests aft er 12 months of treatment with anti depressant medication. All tests were independently administered by two senior att ending doctors. Their inter-rater reliability for the various scales was good； the intraclass coeffi cient （ICC）for NNSE was 0.84， that for the digit span tests was 0.85， that for the TNT-A and TNT-B tests was 0.85， and that for the GDS was 0.88.

In the morning of the day of enrolment， two milliliters of blood was drawn from all participants.These samples were stored at -80 °C and the TIANamp Genomic DNA Kit （DP304） was used for DNA extraction to identify the epsilon 4 isoform of the APOE gene. The TaKaRa LA Taq with GC Buff er （DRR02AG） and BIO-RAD NJ Nini Opti con Real-Time Polymerase Chain Reaction（PCR） System was used for the extracti on， the PCR primer was made by Invitrogen （Shanghai）， and the PCR sequencing was completed by the Shanghai Najorbio Bio-Pharm Technology Company.

The depressed pati ents were treated according to the Treatment and Preventi on Guidelines for Depressive Disorder of China （issued by the Chinese Nedical Associati on in 2006）.[12]All patients were treated with selecti ve serotonin reuptake inhibitors（SSRIs） at adequate dosages for suffi cient ti me. Aft er initial titration of the dosage， the dosages remained stable throughout the acute， continuation and maintenance phases of treatment. The first choice of SSRI was not effective after 6 weeks of treatment（i.e.， decrease in GDS ＜50%） in 19 of the 64 （29.7%）pati ents so they were converted to a different SSRI；9 of these patients （14.1%） had sti ll not improved at the end of the 12 months of follow-up. The first line medicati ons used were sertraline （50 to 150 mg/d）or citalopram （20 to 60 mg/d） and the second-line medicati on was paroxeti ne （20 to 40 mg/d）. Aft er 12 months most pati ents had completed their full course of anti depressant treatment， only 21 of the 64 pati ents（32.8%） were still using anti depressant medication at the 12-month follow-up.

This study was approved by the ethical review committee of the Third Fospital of Fuzhou.

2.3 Stati sti cal analysis

All analyses were conducted using SPSS （version 17.0）soft ware. Chi-squared test， t-test， repeated measures ANOVA， and Spearman correlation coefficients were used for data analysis. All tests were two-tailed and stati sti cal signif i cance was set as p＜0.05.

3. Results

3.1 Comparison of cognitive functioning at baseline

As shown in Table 1， compared to the control subjects the depressed patients had significantly lower total scores on the NNSE， shorter digit spans and took longer to compete the trail making tests. Thus， the depressed patients were more cognitively impaired than control subjects of the same age and gender.

3.2 Comparison of depressive symptoms and cogniti ve functioning in depressed elderly with and without MCI before and aft er treatment

?

As shown in Table 2， the severity of depression at baseline was not stati sti cally diff erent between elderly depressed subjects with and without concurrent NCI.Aft er 12 months of treatment for depression the severity of depression improved signif i cantly in both types of pati ents but those with concurrent NCI at baseline had significantly more residual symptoms of depression aft er 12 months of treatment than pati ents who did not have concurrent NCI at baseline. The magnitude of the change in the before versus aft er GDS score was smaller in the NCI group， but this diff erence did not reach stati sti cal signif i cance （p=0.065）.

With the excepti on of the TNT-A test， all of the cogniti ve measures assessed showed significantly greater impairment in the NCI group at baseline and these diff erences persisted aft er 12 months of treatment for depression. In most cases the individual cognitive tests improved over the 12 months in both the NCI group and the non-NCI group， but the improvement in the backward digit span for the NCI group did not reach statistical signif i cance， the improvement in the forward digit span for the non-NCI group did not reach statistical significance，and the total NNSE score in the non-NCI group showed a non-signif i cant decrease over the 12 months. The improvement in the NNSE score over the 12 months was significantly greater in the NCI group than in the non-NCI group， but there were no significant differences in the magnitude of the improvement between the two groups in any of the other cogniti ve measures.

3.3 Correlati on between severity of depression and cognitive functioning before and after treatment

As shown in Table 3， prior to treatment the severity of depression was only weakly correlated with the severity of cognitive impairment as measured by the diff erent scales.Twelve months aft er initi ati ng anti depressant therapy the residual level of depressive symptoms was significantly associated with the severity of three of the fi ve cogniti ve measures （NNSE， forward digit span and backward digit span）.

3.4 Prevalence of epsilon 4 allele of the APOE genotype and its relationship to the severity of depression and cognitive functioning before and aft er treatment in depressed subjects

Among the depressed patients 14.1% （9/64） had the epsilon 4 allele of the APOE genotype but only 3.3%（1/30） of the control subjects had this allele （Fisher’s Exact Test， p=0.158）. Table 4 compares the depressive and cogniti ve symptoms between the 9 patients with geriatric depression who had the epsilon 4 allele and the 55 patients who did not have the allele. At baseline and at the 12 month follow-up the severity of depressive symptoms between those with and without the APOE epsilon 4 allele was similar. Both types of patients showed significant improvements in depressive symptoms with treatment； the magnitude of improvement was slightly greater in those without the allele but this difference was non-significant （p=0.164）.

At baseline the NNSE and digit span tests results were not significantly different between those with and without the APOE epsilon 4 allele， but those with the allele took significantly longer to complete the TNT-A and TNT-B tests. At the 12-month follow-up all cognitive tests improved in the depressed patients who did not have the APOE epsilon 4 allele， but in the subgroup with the epsilon 4 allele only the two trailmaking tests showed significant improvement； the two digit span tests in those with the epsilon 4 allele showed non-significant improvement and the NNSE showed non-significant deterioration. At the end of the 12 months the NNSE score， the backward digit test and the TNT-A test were significantly worse in the subgroup with the epsilon 4 allele than in the subgroup without the allele. The magnitude of improvement in the NNSE score over the 12 months was significantly greater in those without the allele， but there were no significant differences in the magnitude of improvement between the two subgroups for any of the other cognitive measures.

?

?

4. Discussion

4.1 Main findings

We found that patients who meet criteria for geriatric depression have substantial impairments in their cognitive functioning as assessed by the NNSE， digit span tasks and the Trail-Naking tests. Using the criteria specif i ed by Peterson[8]48% of them （31/64） met criteria for NCI. These findings are similar to those of other studies that find impairments in memory and executive function in elderly individuals with depression，[13，14]which may be related structural and functional impairments in the hippocampus in persons with depression.[15]

We also found that both depressive and cognitive symptoms of elderly individuals with depression improve with standardized treatment for depression with SSRIs. Thus our initial hypothesis that cognitive symptoms in depressed elderly get worse and may progress to full-criteria dementia during anti depressant treatment was not supported， at least for the first year after initiation of anti depressant treatment. When subdividing the depressed group into those with and without NCI at baseline， there were more residual depressive and cognitive symptoms in the NCI group than in the non-NCI group aft er treatment， but the magnitude of the before-versus-after improvement in the NNSE score （the most comprehensive measure of cognitive functioning used in the study） was actually greater in the NCI group. This suggests that the cognitive symptoms in depressed elderly have both trait （i.e.， long-lasting symptoms that may be markers of subsequent dementia） and state （i.e.， symptoms that are amenable to treatment） components. Our findings are similar to those of Yuan and colleagues[16]who report that the responsiveness of depressive symptoms to anti depressant treatment is less marked in depressed elderly with prominent cognitive symptoms， and to those of Zhang and colleagues[17]who report parti al improvement of cogniti ve symptoms in depressed elderly during anti depressant treatment.A study by Butt ers and colleagues[18]also found parti alimprovement in the comorbid cognitive symptoms of depressed elderly； but their study suggested that problems with executive functioning in these patients were more likely to be treatment resistant， a finding that we did not confirm.

Table 4. Comparison of the depressive and cognitive measures between elderly depressed patients with and without the ApoE4 genotype （mean [SD]）

The patients with geriatric depression were more likely to have the APOE epsilon 4 allele than individuals in the control group （a non-significant difference）， but there were only 9 patients with the allele so our results must be considered preliminary， needing confirmation in larger samples that have more patients with the APOE epsilon 4 allele. Our results suggest that some cognitive functions are more severely impaired in depressed elderly who have the epsilon 4 allele than in depressed elderly without the allele both before and aft er anti depressant treatment. Noreover， the NNSE score in pati ents with the APOE epsilon 4 allele decreased over the year of treatment while the corresponding score in patients without the allele increased. The APOE epsilon 4 allele is considered a predisposing gene for sporadic AD，[5]so the presence of this allele in individuals with geriatric depression may mean that more of the comorbid cognitive symptoms are ‘trait’symptoms that are not amenable to anti depressant treatment

4.2 Limitati ons

The patients identified in this study were those who came to a psychiatric outpatient center for treatment， so they were probably more severely depressed than most elderly depressed individuals in the community. We only used a small number of cognitive tests and did not conduct neuroimaging tests so we were limited in our ability to interpret the underlying pathophysiology related to our findings. Nost of the depressed patients had completed their course of treatment at the ti me of the 12-month follow-up so the post-treatment results combined those of patients who were and were not currently taking anti depressants. The sample size was suffi cient to compare cognitive measures between depressed patients and normal controls， and to compare depressive symptoms and cognitive functioning before and after treatment in the depressed pati ents， but the analyses comparing the outcomes for depressed patients who did and did not have NCI and for patients who did and did not have the APOE epsilon 4 allele was inconclusive because of the small number of depressed patients in the subgroups.The analyses comparing these subgroups in which there were significant differences are valid， but the analyses in which there were non-significant differences between subgroups may have been due to Type II errors.

4.3 Significance

Individuals who meet criteria for geriatric depression treated in outpatient psychiatric clinics in China have substantial cognitive def i cits compared to elderly controls without depression； almost one-half of these individuals met criteria for mild cognitive impairment （NCI）. The severity of depression in depressed geriatric patients with and without comorbid NCI are similar before treatment and both subgroups of patients improve significantly with anti depressant treatment， but those with comorbid NCI have more residual depressive symptoms at the end of treatment. The cognitive symptoms in both groups also improve during treatment for depression； in most cases the improvement in cognitive symptoms is greater in the patients with comorbid NCI but at the end of one year patients who had NCI at baseline still have signif i cantly greater cognitive def i cits than those who did not have NCI at baseline.

Only 9 of the depressed elderly patients in this study had the APOE epsilon allele so it is not possible to come to any def i nitive conclusions about its role in depression and dementia. Nevertheless， despite depressive symptoms of similar severity in patients with and without the APOE epsilon allele， several of the cognitive measures assessed were significantly worse in patients with the allele both before and after anti depressant treatment and the magnitude of improvement in the NNSE （the main measure of cognitive functioning） with treatment was significantly smaller in those with the allele.

Conf l ict of Interest

The authors report not conflict of interest related to this manuscript.

Acknowledgement

Authors would like to thank Professor Chen Wei from the Run Run Shaw Fospital， Zhejiang University， for his valuable comments.

Funding

The current study is funded by the Fuzhou ministry of technology general research funds （2010C33015）.

1. Wu LY， Wei J， Li SW. Research progress of depressive disorder with cognitive dysfunction. Chinese Journal of Psychiatry 2004； 37（3）: 188-189. （in Chinese）

2. Fe Y. Impact of geriatric depression on the course of mild cognitive impairment. Chinese Journal of Gerontology 2009；8（29）: 2127-2129. （in Chinese）

3. Lee JS， Potter GG， Wagner FR，Welsh-Bohmer KA， Steff ens DC. Persistent mild cognitive impairment in geriatric depression. Int Psychogeriatr 2007； 19（1）: 125-135.

4. Steff ens DC， Plassman BL， Felms NJ， Welsh-Bohmer KA，Saunders AN， Breitner JC. A twin study of late-onset depression and apolipoprotein E epsilon 4 as risk factors for Alzheimer’s disease. Biol Psychiatry 1997； 41（8）: 851-856.

5. Zubenko GS， Fenderson R， Sti ffl er JS Stabler S， Rosen J，Kaplan BB. Association of the ApoE epsilon 4 allele with clinical subtypes of late life depression. Biol Psychiatry 1996； 40: 1008-1016.

6. Chinese Society of Psychiatry， Chinese Nedical Associati on.Chinese Classification of Mental Disorders Third Edition（CCND-3）. Jinan: Shangdong Science and Technology Press；2001.

7. Sheikh JI， Yesavage JA， Brooks JO， III， Friedman LF， Gratzinger P， Fill RD， et al. Proposed factor structure of the Geriatric Depression Scale. Int Psychogeriatr 1991； 3: 23-28

8. Petersen RC. Nild cognitive impairment as a diagnostic entity. J Intern Ned 2004； 256: 183-194.

9. Reisberg B， Ferris SF， de Leon NJ， Crook T. Global Deterioration Scale （GDS）. Psychopharmacol Bull 1998；24（4）: 661-663.

10. Folstein NE， Folstein SE， NcFugh PR. “Nini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975； 12（3）: 189-198.

11. Reitan RN. Validity of the Trail Naking Test as an indicator of organic brain damage. Percept Mot Skills 1958； 8: 271-276.

12. Chinese Society of Psychiatry， Chinese Nedical Associati on.China's Guidelines for the Treatment and Preventi on of Depression. Chinese Nedical Associati on； 2006. （in Chinese）

13. Taragano FE， Bagnatti P， Allegri RF. A double-blind，randomized clinical trial to assess the augmentati on with nimodipine of anti depressant therapy in the treatment of vascular depression. Int Psychogeriatr 2005； 17（3）: 487-498.

14. Na X， Wang JF， Bao F， Zhu F， Li ZJ. Auditory event-related potential P300 in geriatric depression and dementi a.Chinese Journal of Geriatrics 2000； 19（6）: 411-413. （in Chinese）

15. Xu L， Xu J. Research progress of brain NRI in geriatric depression. Chinese Journal of Psychiatry 2009； 42（3）: 182-184. （in Chinese）

16. Yuan YG， Ye Q， Li FL， Wu RZ， Lu R， Chen F. Clinical feature of depression in senior people with cognitive dysfunction.Chinese Journal of Clinical Rehabilitati on 2005； 9（32）: 26-28.（in Chinese）

17. Zhang LL， Zheng FB， Deng Y， Fe ZG， Na JS， Li JF. An analysis on the cognitive function in depression before and aft er treatment. Journal of Neuroscience and Mental Health 2007； 7（4）: 262-265. （in Chinese）

18. Butt ers NA， Becjer JT， Nebes RD， Zmuda ND， Nulsant BF，Pollock BG， et al. Changes in cognitive functioning following treatment of late-life depression. Am J Psychiatry 2000；157（12）: 1949-1954.

老年抑郁癥患者抗抑郁藥治療期間認(rèn)知與抑郁癥狀的變化及其與ApoEε4等位基因的關(guān)系

陳浙麗1* 蘭光華2沈鑫華1潘新根1陳小鄖1李建華1

1浙江省湖州市第三人民醫(yī)院老年精神科 浙江湖州
2江蘇省蘇州大學(xué)附屬第一醫(yī)院精神科 江蘇蘇州

22009061@qq.com

背景老年抑郁癥伴有的認(rèn)知損害可能不隨抗抑郁治療而好轉(zhuǎn)，可能會(huì)進(jìn)展為癡呆。目的探討老年抑郁癥患者抗抑郁治療前后認(rèn)知和抑郁癥狀變化的關(guān)系及其與ApoEε4等位基因的關(guān)聯(lián)。方法檢測(cè)64例首發(fā)老年抑郁癥患者和31名非抑郁老年人的ApoEε4等位基因。抑郁癥患者接受標(biāo)準(zhǔn)化的選擇性5-羥色胺再攝取抑制劑（Selective Serotonin Reuptake Inhibitors, SSRIs）治療，在基線和治療12個(gè)月后分別對(duì)研究對(duì)象進(jìn)行老年抑郁量表(The Geriatric Depression Scale, GDS)、簡(jiǎn)易智能狀態(tài)檢查（Mini-Mental State Examination,MMSE）、數(shù)字廣度測(cè)驗(yàn)（digit span task）和連線測(cè)驗(yàn)A, B（Trail making test, TMT-A, TMT-B)的測(cè)評(píng)。結(jié)果與對(duì)照組比較，治療前老年抑郁癥組患者的認(rèn)知功能較差，其中31（48%）例符合輕度認(rèn)知功能損害（mild cognitive impairment, MCI）的標(biāo)準(zhǔn)。治療后，無論共病MCI與否，抑郁癥患者的認(rèn)知和抑郁癥狀均有改善，但伴MCI者殘留癥狀更多。基線時(shí)認(rèn)知癥狀的嚴(yán)重程度與抑郁癥狀的嚴(yán)重程度不存在相關(guān)性，但在12個(gè)月隨訪時(shí)二者存在正相關(guān)?；颊咧杏?4%（9/ 64）攜帶ApoEε4等位基因，而對(duì)照中僅有3%（1/31）攜帶APOEε4等位基因（Fisher精確檢驗(yàn)，p=0.158）。與未攜帶ApoEε4等位基因的抑郁癥患者相比，攜帶此等位基因的抑郁癥患者治療前、后均有較嚴(yán)重的認(rèn)知癥狀，但僅某些方面的差異具有統(tǒng)計(jì)學(xué)意義。結(jié)論老年抑郁癥患者普遍存在認(rèn)知損害。經(jīng)標(biāo)準(zhǔn)的SSRI治療后，患者的抑郁和認(rèn)知癥狀均有所改善，但共病MCI的患者治療后殘留更多的抑郁和認(rèn)知癥狀。老年抑郁癥患者中攜帶ApoEε4等位基因與較嚴(yán)重的認(rèn)知損害有關(guān)，可能與抗抑郁劑治療認(rèn)知癥狀療效欠佳有關(guān)。

Background：Co-occurring cogniti ve impairment in geriatric depression may not improve with anti depressant treatment and it may progress to dementia.Aim：Assess the relati onship between changes in cogniti ve and depressive symptoms among patients with geriatric depression and their associati on with the APOE epsilon 4 allele before and aft er anti depressant treatment.Methods：The presence of the APOE epsilon 4 allele was assessed in 64 incident cases of geriatric depression and 31 elderly individuals without depression and the Geriatric Depression Scale （GDS）， Nini-Nental State Examinati on （NNSE）， digit span test， and Trail Naking Tests A and B （TNT-A， TNT-B） were administered to these subjects at baseline and 12 months aft er baseline， during which ti me the depressed group received standardized treatment with selecti ve serotonin reuptake inhibitors （SSRIs）.Results：Prior to treatment patients with geriatric depression had signif i cantly worse cogniti ve functi oning than control subjects and 31 （48%） met criteria for mild cognitive impairment （NCI）. Aft er treatment depressed pati ents with and without comorbid NCI both had signif i cant improvements in their depressive and cognitive symptoms， but those with NCI had more residual symptoms. The severity of cogniti ve symptoms was not associated with the severity of depressive symptoms at baseline， but they were positi vely correlated at the 12-month follow-up. The APOE epsilon 4 allele was identified in 14% （9/64） of the patients and in 3% （1/31）of the controls （Fisher’s Exact Test， p=0.158）. Compared to depressed patients without the allele， depressed pati ents with the allele had more severe cogniti ve def i cits both before and aft er treatment， though only some of these differences were statistically significant.Conclusions：There is substanti al cognitive impairment in elderly individuals with geriatric depression. Both the depressive and cognitive symptoms improve with standard SSRI treatment， but individuals with comorbid NCI have more residual depressive and cognitive symptoms aft er treatment. The APOE epsilon 4 allele is associated with greater cognitive impairment in geriatric depressed patients and may be associated with less responsiveness of cognitive symptoms to anti depressant treatment.

10.3969/j.issn.1002-0829.2013.02.006

1Department of Geriatric Psychiatry， Third People’s Fospital of Fuzhou， Fuzhou City， Zhejiang Province， China

2Psyciatry Department， The First Affiliated Fospital of Soochow University， Suzhou， Jiangsu Province， China

Author’s correspondence: 22009061@qq.com

（received: 2012-11-12； Accepted: 2013-1-10）

Dr. Zheli Chen graduated from Wenzhou Medical College in 2002 and received a Master's of Medicine degree from Suzhou University in 2012. She has worked at the Third Hospital of Huzhou since 2002 in the Department of Geriatric Psychiatry where she is currently an attending psychiatrist. Her main research interests are the clinical characteristics and cognitive functioning of geriatric depression and the early identifi cati on and treatment of Alzheimer's Disease.