680

Depression Scores Following Migraine Treatment in Patients Attending a Specialized Center for Headache and Neurology

Héctor Miranda, MD; Gilberto Ortiz; Sandra Figueroa; Cynthia M. Pérez, PhD; Erick Suárez, PhD

Objective.—To determine the changes in clinical characteristics and depression levels among patients follow-ing treatment for migraine.

Background.—Epidemiologic studies have provided consistent evidence regarding an association betweenmigraine and depression. In Puerto Rico, however, migraine has not yet been systematically investigated.

Methods.—A chart review of 144 Puerto Rican patients who presented with migraine, diagnosed according tothe International Headache Society criteria, and depression over a 2-year period was performed. The level of de-pression, before and after migraine treatment, was evaluated using the Zung Self-rating Depression Scale.

Results.—The mean age of patients was 37.0

�

14.4 years; 77.1% were women. More than half (52.8%) re-ported severe headache and 56.9% reported a monthly frequency of five attacks or more. Nearly 9% were usingantidepressant therapy and 8% were under psychiatric treatment. The mean Zung index score at baseline was 50.6

�

10.9. Following treatment with triptans, the intensity and frequency of migraine and the Zung index score de-creased significantly (

P

�

.00001). A trend for a greater reduction in Zung index scores among patients receivingtriptan medications for more than a year was demonstrated (

P

�

.07).Conclusions.—These results indicate that migraine treatment with triptans appears to be effective in decreas-

ing the headache frequency and intensity, and depression levels, independent of antidepressant medication use orpsychiatric treatment.

Key words: migraine treatment, depression, headache, headache center, headache treatment

(

Headache

2001;41:680-684)

Multiple epidemiologic studies have reported anassociation between migraine and psychiatric condi-tions, specifically depression.

1-15

During the last fewyears, the link was explored from different points ofview. Some authors

16-18

explored the possibility of acommon etiologic factor between migraine and depres-sion. Others suggested a bidirectional link betweenmigraine and depression in which the presence of one

of the conditions predispose to the development ofthe other.

19

Alternatively, the theory that migraine or

any severe headache

predisposes to the developmentof depression

20

is gaining popularity.In this study, we examined the suggested unidi-

rectional relationship between migraine and depres-sion in which the presence of migraine predisposesto the development of depression. We performed achart review of 144 patients who presented with mi-graine and depression on the initial evaluation at theCenter for Headache Management at the San Fran-cisco Hospital in San Juan, Puerto Rico. The diagno-sis of migraine was according to the InternationalHeadache Society criteria, and the intensity of head-ache was graded on a numeric scale in which 4 is thehighest level of pain and 0 is no headache. The di-agnosis of depression was made using the Zung Self-rating Depression Scale. Subsequently, these patients

From the Specialized Headache Management and NeurologyCenter, San Francisco Hospital (Dr. Miranda, Mr. Ortiz, andMs. Figueroa); and the Department of Biostatistics and Epide-miology (Drs. Pérez and Suárez) and Neurology (Dr. Miranda),Medical Sciences Campus, University of Puerto Rico, San Juan.

Address all correspondence to Dr. Hector S. Miranda-Del-gado, CEDOC Headache Center, San Francisco Tower, Suite409, 365 De Diego Avenue, San Juan, PR 00923.

Accepted for publication March 12, 2001.

Headache

681

were reevaluated to determine changes in depressionlevels.

SUBJECTS AND METHODS

All patients with a diagnosis of migraine seen atthe Center for Headache Management between 1996and 1998 were evaluated for the presence of depres-sion. During their initial evaluation, a self-adminis-tered questionnaire was used to determine the levelof depression, if any, using the Zung scale. After thisevaluation, the patients were contacted by telephonein December 1998 and January 1999 to determine thelevel of depression using the same scale after at least6 months of treatment for migraine.

Frequency distributions were computed for gender,headache severity, frequency of attacks, antidepres-sant drug use, and psychiatric treatment. Mean

�

SD(range) was calculated for each of the following vari-ables: age and EAD index. The EAD index was de-fined as the Zung score divided by 0.80.

21

An EAD in-dex of more than 55 in patients older than 60 years oran index of more than 50 in patients aged 60 years oryounger was suggestive of depression.

21

Bivariate anal-ysis consisted of the McNemar

�

2

test and paired

t

testfor categorical and continuous variables, respectively.

22

Matched odds ratios and 95% confidence intervals(CI) were calculated as a measure of association be-tween categorical variables.

23

Using the

�

2

test, a strati-fied analysis by the time period elapsed between thetwo scale administrations was performed (

�

12 monthsversus

�

12 months). The proportions of change of thefollowing parameters were recorded: severity, monthlyfrequency, antidepressant drug use, psychiatric treat-ment, and presence of depression according to theZung scale. Finally, the Kruskal-Wallis test was usedto evaluate the change in the EAD index (administra-tion 2 score minus administration 1 score) during thetime elapsed between the two scale administrations.

24

Data entry and analyses were performed with Epi-Info, Version 6.04b.

25

RESULTS

The mean age of patients was 37.0

�

14.4 years.Seventy-seven percent of the patients were women(Table 1). During the initial administration, more

than half (52.8%) of the patients suffered severeheadaches (4 using the numeric scale); nearly 57%reported a monthly frequency of five attacks or more;8.3% reported antidepressant therapy and 7.6% wereunder psychiatric treatment. The mean EAD indexwas 50.6

�

10.9 (range, 28 to 83) (data not shown).The time elapsed between the two administrationswas less than 12 months in 57.6% of patients.

The odds of classifying a severe headache duringthe first administration as nonsevere during the sec-ond administration were nearly three times (95% CI,1.65 to 5.12) the odds of classifying a nonsevere head-ache during the first administration as severe duringthe second administration (Table 2). The odds of suf-fering five or more monthly headache attacks duringthe initial administration to less than five monthlyheadache attacks during the second administrationwere nearly 19 times (95% CI, 5.64 to 74.62) the oddsof less than five monthly headache attacks during thefirst administration to five or more monthly headacheattacks during the second administration. The oddsof antidepressant therapy use during the first admin-istration to nonuse during the second administrationwere five times (95% CI, 0.15 to 1.58) the odds of an-tidepressant therapy nonuse during the first adminis-

Table 1.—Frequency Distribution of Baseline Characteristics in Study Population (N

�

144)

VariableNo. (%)

of Patients

Women 111 (77.1)Age

�

35 y 72 (50.0)Time period between two administrations, y

�

1 83 (57.6)

�

1 61 (42.4)Headache intensity

Mild/moderate 68 (47.2)Severe 76 (52.8)

Monthly frequency of headache, attacks

�

5 62 (43.1)

�

5 82 (56.9)Antidepressant therapy

Use 12 (8.3)Nonuse 132 (91.7)

Psychiatric evaluationTreatment 11 (7.6)No treatment 133 (92.4)

682

July/August 2001

tration to use during the second administration. Thisreduction was not statistically significant (

P

�

.10).The odds for psychiatric treatment during the firstadministration to no psychiatric treatment during thesecond administration were 5.6 times (95% CI, 0.16to 1.81) the odds of no psychiatric treatment duringthe first administration to psychiatric treatment at thesecond administration. This reduction was not statis-tically significant (

P

�

.10). The odds of classifying apatient as depressed using the Zung scale during thefirst administration as nondepressed during the sec-ond administration were 13 times (95% CI, 3.76 to51.39) the odds of classifying a patient as nonde-

pressed during the first administration as depressedduring the second administration.

The mean change in the EAD index between bothadministrations was

5.82

�

0.82 units. That is, on av-erage, the EAD index decreased 5.8 units between thetwo administrations. This change was statistically sig-nificant by the Kruskal-Wallis test (

P

�

.00001) (datanot shown).

Analysis of these parameters by the time elapsedbetween administrations revealed no significant changes(

P

�

.10) in intensity, monthly frequency, antidepressanttherapy use, psychiatric treatment, and presence of de-pression (Table 3). When the mean change in the

Table 2.—Matched Odds Ratios for Changes in Clinical Parameters

Change in Clinical Parameter Odds Ratio 95% Confidence Interval

Severe to nonsevere headache during initial evaluation versus nonsevere to severe headache during final evaluation (n

�

144) 2.89 1.65 to 5.12Monthly attack frequency

�

5 attacks to

�

5 attacks during initial evaluation versus

�

5 attacks to

�

5 attacks during final evaluation (n

�

144) 18.67 5.64 to 74.62Antidepressant use to no use during initial evaluation versus no antidepressant

therapy use to use during final evaluation (n

�

143) 5.0 0.15 to 1.58Need for psychiatric evaluation to no need during initial evaluation versus no need to

need for psychiatric evaluation during final evaluation (n

�

143) 5.6 0.16 to 1.81Presence of depression to absence during initial evaluation versus absence of

depression to presence during final evaluation (n

�

137) 12.67 3.76 to 51.39

Table 3.—Stratified Analysis of Changes in Clinical Parameters by Time Elapsed Between the Two Administrations

Change

Time Elapsed Between Administrations

�

12 months

�

12 months

P

Severe intensity in administration 1 to nonsevere intensity in administration 2 68.8 67.9

�

.10*Monthly headache frequency

�

5 attacks in administration 1 to

�

5 attacks in administration 2 69.4 66.7

�

.10*Antidepressant therapy use in administration 1 to nonuse in administration 2 40 42.9

�

.10*Need for psychiatric evaluation in administration 1 to no need in administration 2 50 42.9

�

.10*Presence of depression in administration 1 to no depression in administration 2 51.4 73.1

�

.10*Mean change (

�

SE) in EAD index between administrations

5.10

�

1.15(n

�

83)

7.89

�

1.09(n

�

53).074†

Values are percentages unless otherwise indicated.*

P

obtained from

�

2

test.†

P

obtained from Kruskal-Wallis test.

Headache

683

EAD index between administrations was analyzed,there was a trend for a greater reduction among pa-tients who experienced a longer time period betweenthe two administrations (

P

�

.074).

COMMENTS

The mean changes in the EAD index observed be-tween the two administrations were statistically signifi-cant, suggesting that the depression score decreasedover time. Significant changes in the headache inten-sity and monthly frequency were also observed.

Prior studies that evaluated the association be-tween migraine and depression suggest that the linkbetween these conditions is related to a common ori-gin or etiology. This observation was based on thefact that the tricyclic antidepressant drugs are effec-tive in both conditions and the presence of serotoninand adrenaline as a common neurotransmitter. Inthis study, we used the unidirectional approach inwhich the presence of one of the conditions predis-posed to the development of the other. Using theEAD index, we obtained values that point to the factthat the depression score that was identified on theinitial evaluation decreased in a direct relationshipwith the improvement in the frequency and intensityof the migraine attacks.

The Zung scale is a questionnaire that enables apatient to auto-evaluate his or her levels of depres-sion. In this study, during the first evaluation, pa-tients answered the questionnaire in the Center.In the reevaluation, the questionnaire was adminis-trated by telephone, and even though patients did notdirectly complete the scale, the auto-evaluation of de-pression levels was obtained. Even though we recog-nize that the Zung Scale alone is not adequate for aconclusive diagnosis of depression, the informationgiven by this scale suggests the level of depression, ifany, that a patient has at a given time and allows oneto know if this level had increased or decreased.

In this study, we did not control the time betweenthe evaluations, the compliance with the treatment,the levels of depression, or the type of medicationsused to improve the migraine events. In a futurestudy, we need to control those parameters to evalu-ate if the relationship suggested by this study is main-tained with different patterns of frequency and inten-

sity of the attacks. Also, we need to account for otherfactors that influence the presence of depression suchas the type of treatment that the patient received fordepression or the use of alternative methods to im-prove the depression such as herbs or relaxation tech-niques. Controlling these parameters will permit usto know if the management of the headache de-creases depression levels in a direct proportion as issuggested by this study.

Acknowledgment: Appreciation is expressed to Pe-

tra Burke, MD, Professor, Department of Neurology,

University of Puerto Rico School of Medical Sciences, for

her support and revision of this investigation work.

REFERENCES

1. Garvey MJ, Tollefson GD, Schaffer CB. Migraineheadaches and depression.

Am J Psychiatry.

1984;141:986-988.

2. Holroyd, KA. Recurrent headache. In: Holroyd KA,Creer TL eds. Self-Management of Chronic Disease:Handbook of Clinical Interventions and Research.Orlando, Fla: Academic Press; 1986:373-413.

3. Davis RA, Wetzel RD, Kashiwagi MT, McClure JNJr. Personality, depression and headache type.

Head-ache.

1976;16:246-251.4. Kashiwagi T, McClure JN Jr, Wetzel RD. Headache

and psychiatric disorders.

Dis Nerv Syst.

1972;33:659-663.

5. Weatherhead AD. Psychogenic headache.

Head-ache.

1980;20:47-54.6. Wolff HG. Headache and Other Head Pain. 2nd ed.

New York: Oxford University Press; 1963.7. Ziegler DK, Rhodes RJ, Hassanein RS. Association

of psychological measurements of anxiety and de-pression with headache history in a non-clinic popu-lation.

Res Clin Stud Headache.

1978;6:123-135.8. Andrasik F, Blanchard EB, Arena JG, Teders SJ,

Teevan RC, Rodichok LD. Psychological function-ing in headache sufferers.

Psychosom Med.

1982;44:171-182.

9. Cadoret RJ, Widmer RB, Troughton EP. Somaticcomplaints—harbinger of depression in primary care.

J Affective Disord.

1980;2:61-70.10. Cassidy WL, Flanagan NB, Spellman ME. Clinical

observation in manic-depressive disease.

JAMA.

1957;164:1535-1546.

11. Crisp AH, Kalucy RS, McGuiness B, Ralph PC,Harris G. Some clinical, social, and psychological

684

July/August 2001

characteristics of migraine subjects in the generalpopulation.

Postgrad Med J.

1977;53:691-697.12. Puca F, Genco S, Savarese M, et al. Stress, depres-

sion, and anxiety in primary headache sufferers:evaluation by means of the SCL-90-R.

Headache Q.

1992;3:187-192.13. Shulman BH. Psychiatric approach to the patient

hospitalized for headache.

Headache Q.

1993;4:63-67.14. Blanchard EB, Kirsch CA, Appelbaum KA, Jaccard

J. The role of psychopathology in chronic headache:cause of effect?

Headache.

1989;29:295-301.15. Breslau N, Davis GC. Migraine, major depression

and panic disorder: a prospective epidemiologic studyof young adults.

Cephalalgia.

1992;12:85-90.16. D’Andrea G, Welch KM, Riddle JM, Grunfeld S, Jo-

seph R. Platelet serotonin metabolism and ultrastruc-ture in migraine.

Arch Neurol.

1989;46:1187-1189.17. Meltzer HY, Lowy MT. The serotonin hypothesis of

depression. In: Meltzer HY, ed. Psychopharmacol-ogy: The Third Generation of Progress. New York:Raven Press; 1987:513-526.

18. Glover V, Jarman J, Sandler M. Migraine and de-

pression: biological aspects.

J Psychiatr Res.

1993;27:223-231.

19. Breslau N, Davis GC, Schultz LR, Peterson EL. Mi-graine and major depression: a longitudinal study.

Headache.

1994;34:387-393.

20. Rueslan N, Schultz L, Welch KM. Is comorbidity withmajor depression specific to migraine? In: Programand Abstracts of the 12th Migraine Trust Interna-tional Symposium. September 1-4, 1998; London, UK.

21. McDowell I, Newell C. Measuring Health: A Guideto Rating Scales and Questionnaires. 2nd ed. NewYork: Oxford University Press; 1996.

22. Rosner B. Fundamentals of Biostatistics. 4th ed.Belmont: Duxbury Press; 1995.

23. Gordis L. Epidemiology. Philadelphia, Pa: WB Saun-ders Co; 1996.

24. Daniel WW. Applied Nonparametric Statistics. 2nded. Boston, Mass: PSW-KENT Publishing Co; 1990.

25. Dean AG, Dean JA, Burton AH, Dicker RC. EPI-INFO, Version 6: A Word Processing, Database,and Statistics Program for Epidemiology on Micro-computers. Stone Mountain, Ga: USD, Inc; 1990.