Analysis of Leukocytes in Medication-Overuse Headache, Chronic Migraine, and Episodic Migraine

JOBNAME: No Job Name PAGE: 1 SESS: 51 OUTPUT: Thu Apr 28 17:06:53 2011/v2451/blackwell/journals/head_v0_i0/head_1902

Research Submission

Analysis of Leucocytes in Medication-Overuse Headache,Chronic Migraine, and Episodic Migrainehead_1902 1..12

Cassiano M. Forcelini, MD; Denise C. M. Dantas, PhD; Clarice Luz, PhD; Ricardo Santin, MD;Airton T. Stein, MD, PhD; Helena M. T. Barros, MD, PhD; Liselotte M. Barea, MD, PhD

Objective.—To evaluate the number of immune cells in the peripheral blood of medication-overuse headache (MOH),chronic migraine (CM), and migraine without aura (MWA) patients, as well as from controls.

Background.—Migraine has been linked to immunologic disturbances, but the role of the immune system in chronic formsof headache that evolve from migraine has not been studied. Psychiatric co-morbidity has been related to both headachechronification and immunologic alterations.

Methods.—This cross-sectional study comprised 68 subjects divided in 4 groups: MOH, CM, MWA, control. Subjects weregender-matched, had no physical co-morbidity, and were taking only acetaminophen. Clinical and psychological data wererecorded in a standardized protocol. Samples of peripheral blood for hematological analysis were obtained in the morningduring the ictal (MOH, CM, and MWA groups) and interictal periods (MWA group), as well from control group.

Results.—A higher lymphocyte count was measured in MOH patients relative to the MWA patients (mean � standarddeviation: 2448.7/mm3 � 775.8 vs 1859.7/mm3 � 564.7; P = .027). The numbers of blood lymphocytes for CM and controlsubjects were 2086.1/mm3 � 540.5 and 1961.7/mm3 � 385.6, respectively. Multiple linear regression analysis demonstrated thatonly MOH and MWA groups remained associated with lymphocyte count (B = 540.7; CI 95%: 55.2-1026.1; P = .03; R2 = 19.2%).Analysis for linearity of variables in the spectrum control/MWA/CM/MOH resulted positive for body mass index (from23.5 � 3.25 in controls to 26.5 � 4.49 in MOH patients; P = .034), scores on Beck Depression Inventory (from 3.29 � 3.05 to14.65 � 11.21; P < 0.001) and Hamilton Anxiety Scale (from 4.29 � 3.93 to 23.24 � 11.01; P < 0.001), hemoglobine (from13.7 � 0.79 to 14.6 � 1.31; P = .022), and lymphocyte count (from 1961.7 � 385.6 to 2448.7 � 775.8; P = .01), but negative forCD8+ T lymphocytes (from 34.0 � 8.82 to 30.0 � 6.64; P = .046).

Conclusions.—A higher lymphocyte count in the MOH group relative to the MWA group may indicate a chronicinflammatory state. Several clinical and laboratorial characteristics have a range along a spectrum extending from healthysubjects to patients suffering from chronic forms of migraine.

Key words: medication-overuse headache, chronic migraine, migraine, leucocyte, lymphocyte, psychiatric co-morbidity

Abbreviations: CM chronic migraine, MOH medication-overuse headache, MWA migraine without aura

(Headache 2011;••:••-••)

From the Programa de Pós-Graduação em Ciências da Saúde, Universidade Federal de Ciências da Saúde de Porto Alegre, PortoAlegre, RS, Brazil (C.M. Forcelini, D.C.M. Dantas, R. Santin, A.T. Stein, H.M.T. Barros, and L.M. Barea); Faculdade de Medicina,Universidade de Passo Fundo, Passo Fundo, RS, Brazil (C.M. Forcelini); Laboratório Vitrus, Porto Alegre, RS, Brazil (C. Luz).

Research support: Dr Barros and Dr Stein are researchers sponsored by Conselho Nacional de Desenvolvimento Científico eTecnológico (CNPq).

Address all correspondence to C.M. Forcelini, Universidade de Passo Fundo, Faculdade de Medicina, R. Teixeira Soares 817Centro, Passo Fundo, Rio Grande do Sul, CEP 99010-080, Brazil.

Accepted for publication February 27, 2011.

Conflict of Interest: Dr Forcelini is principal investigator of a study sponsored by Genzyme. Dr Barea is principal investigator orco-investigator of studies sponsored by Roche, Novartis, and Boehringer Ingelheim. The other authors declare no conflict ofinterest.

ISSN 0017-8748doi: 10.1111/j.1526-4610.2011.01902.xPublished by Wiley Periodicals, Inc.

Headache© 2011 American Headache Society

1

12

3

4

5

6

78

9

101112131415161718192021222324252627282930

31

32

33

34

353637

3839

4041

42

434445

11


Up to 6.9% of the general population is affectedby chronic daily headache,1 represented mostly bychronic migraine and medication-overuse headache.These conditions usually evolve from episodicmigraine and, until recently, were collectively calledtransformed migraine.

The understanding of migraine pathophysiologyhas dramatically improved, but the full picture hasnot been delineated.2 Migraine precipitation involvesthe activation of sensory neurons in the trigeminoc-ervical complex and other brain structures thatparticipate in the modulation of the pain producedby structures of the cranium. Although not demon-strated in humans, there are considerable data sug-gesting that plasma protein extravasation in the duralvessels acts as a form of sensitization that takesplace during migraine.2,3 Such a sterile neurogenicinflammatory model enables the involvement of theimmune system in the precipitation or aggravation ofmigraine. Co-morbidity of migraine and atopicdiseases has been reported, leading to the suspicionthat the immune system may be dysfunctional inmigraineurs.4,5 Accordingly, both humoral and cellularcomponents of the immune system have been studiedin migraine patients since the 1960s, but there aremany conflicting data on this issue.6

Compared to episodic migraine, the neurobiol-ogy of medication-overuse headache and chronicmigraine is poorly understood. Progressive dysfunc-tion of central pain modulation by derangement ofserotonergic7 and neuroendocrine8 systems offers apromising direction for research. Besides, the roleof psychiatric co-morbidity9,10 and obesity11,12 asrisk factors for headache chronification has beenpointed out.

The role of the immune system in medication-overuse headache and chronic migraine has not beenstudied. We hypothesize that the immune system mayplay a role not only in migraine but also, and in anexacerbated manner, in the chronic forms of head-ache that evolve from episodic migraine. Accordingto this theory, the present study aimed to evaluatethe number of immune cells in the peripheral bloodof medication-overuse headache, chronic migraine,and migraine without aura patients, as well as fromcontrols.

PATIENTS AND METHODSPatients.—The study protocol was approved by the

ethics committees of Universidade Federal de Ciên-cias da Saúde de Porto Alegre, Porto Alegre, RS,Brazil and of São Vicente de Paulo Hospital, PassoFundo, RS, Brazil. All patients gave their writtenagreement.

This cross-sectional study (Figure) included 17consecutive medication-overuse headache patients(11 women and 6 men; mean age � SD 38.9 � 12.8years) and the same number of consecutive gender-matched subjects affected by chronic migraine(31.0 � 10.4 years) or migraine without aura(31.4 � 9.7 years) diagnosed according to the Inter-national Classification of Headache Disorders(ICHD-II) and its revised (ICHD-IIR) criteria.13-15

All patients presented for the 1st time at the neurol-ogy outpatient department of the São Vicente dePaulo Hospital, in Passo Fundo, RS, Brazil. All werein the headache (ictal) phase at the time of bloodsample collection, although samples were also col-lected from migraine without aura patients during theinterictal phase (mean of 6.8 days later). Medication-overuse headache and chronic migraine patientsevolved from a retrospective diagnosis of migrainewithout aura months or years previously. Seventeenconsecutive gender-matched controls without head-ache (36.8 � 13.7 years) were recruited among therelatives of other patients who attended the sameoutpatient clinic, but for complaints other than head-ache. The period of parallel recruitment of patientsand controls extended from April 2007 to February2009 (Figure), and a total of 219 subjects werescreened (163 for patients and 56 for controls).

All included subjects had been drug-free (exceptfor acetaminophen) for at least 1 month, had nophysical co-morbidity, and were not smokers. Onlypatients who were taking acetaminophen wereincluded because this analgesic has been widely uti-lized for pain relief and there is no evidence thattherapeutic doses of acetaminophen can modifyimmune parameters like cytokines16 or leucocyteactivation.17

Methods.—Blood samples were obtained via thecubital vein from all subjects in the morning (from7:00 to 11:00 am). Hematological analysis was

2

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92


performed on the same day and comprised a whiteblood cell count, hematocrit, hemoglobin (CellDyn 3000; Abbott, Mountain View, CA, USA), andcytometric analyses (FACSCalibur; Becton Dickin-son, San Jose, CA, USA) with monoclonal antibodies(Becton Dickinson) for T lymphocytes (CD4/CD8/CD3 BD Tritest), B lymphocytes (CD3/CD19Simultest/CD45 PerCP), and natural killer (NK) cells(CD3/CD16+CD56 Simultest).

Clinical data recorded from patient interviewsincluded age, gender, race, weight, height, body massindex, number of acetaminophen pills (500 mg or750 mg) taken per day on the last 4 months, presenceof depression, anxiety, or stress according to theBeck Depression Inventory (1979 revised version),18

Hamilton Anxiety Scale (HAM-A),19 and LippInventory for Stress Symptoms in Adults,20 respec-tively. The cut-off point for defining the presence ofmild depression was a score of at least 10 on the BeckDepression Inventory,21 while the correspondingthreshold for diagnosing mild anxiety was 21 on theHamilton Anxiety Scale.19 Psychometric instrumentsfor assessing the presence of depression, anxiety, andstress were included in the study because it has beenwidely recognized that psychological disturbances areclosely related to migraine and chronic daily head-ache,9,10 as well as to immunologic alterations.22

Statistical Analysis.—The statistical analyses wereperformed with the Statistical Package for the SocialSciences (SPSS) version 12.0. (SPSS Inc., Chicago, IL,

Figure.—Study design flow chart.

Headache 3

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32


USA).One-wayANOVA was used to compare means,and chi-square test with adjusted residual analysis wasused to assess the proportions of stress, depression,and anxiety between the groups.The post hoc analysisincluded Bonferroni and Dunnett T3 tests, the latterfor variables with unequal variances (depression andanxiety scores). To compare means obtained ictallyand interictally from migraine without aura patients,t-test for paired samples was utilized. Sample size wascalculated considering power = 80% to detect a differ-ence of 1 effect size = 1 between groups with a signifi-cant level a of 0.05. The accepted level of statisticalsignificance was P < .05 (2-tailed).

As a secondary analysis, multiple linear regres-sion (enter method) was used for some variables thattheoretically could be related to the variation in lym-phocytes number (dependent variable). The variablesconsidered as predictors were age, body mass index,score on the Beck Depression Inventory, score onthe Hamilton Anxiety Scale, and headache groups.Analysis of variance for linear trend (analysis for lin-

earity) for quantitative variables was also performedafter ordering the 4 headache groups in the followingmanner: control, migraine without aura, chronicmigraine, and medication-overuse headache. Thisorder was based on a theoretical gradient of severity,where subjects without headache fall at 1 extreme ofthe spectrum while patients with daily or near-dailyheadache are clustered at the other end.

RESULTSDemographic and Clinical Characteristics

(Table 1).—Women comprised the majority ofincluded subjects in the 4 clinical groups (11 womenand 6 men in each group). Almost all subjects wereCaucasian (94.1%), reflecting the local ethniccomposition. As expected according to the inclu-sion criteria, the use of analgesics was 6 times higherin medication-overuse headache patients than inchronic migraine patients and 11 times higher than inmigraine without aura patients.

Table 1.—Demographic and Clinical Characteristics of Patients and Control Subjects (Mean � SD)

Control Subjects Migraine Without Aura (Ictal) Chronic Migraine Medication-Overuse Headache

Number of subjects 17 17 17 17Female 11 11 11 11Male 6 6 6 6

Race (%) C: 88.2 C: 100 C: 94.1 C: 94.1Age (years) 36.8 � 13.7 31.4 � 9.7 31.0 � 10.4 38.9 � 12.8Weight (kg) 67.3 � 14.3 65.2 � 10.9 64.2 � 16.9 71.9 � 11.8Height (m) 1.68 � 0.094 1.66 � 0.07 1.61 � 0.08 1.64 � 0.08Body mass index 23.5 � 3.25 23.8 � 3.06 24.2 � 4.92 26.5 � 4.49Depression score 3.29 � 3.05 5.47 � 4.00 8.65 � 6.50† 14.65 � 11.21†Anxiety score 4.29 � 3.93 6.94 � 4.19 15.88 � 11.41‡ 23.24 � 11.01‡Presence of stress (%) 5.9 11.8 64.7§ 70.6§Presence of depression (%) 0 5.9 29.4 52.9¶Presence of anxiety (%) 0 0 23.5 58.8††Acetaminophen pills/month 0.3 � 0.7 3.8 � 2.2 7.2 � 3.4 43.0 � 27.2‡‡

†Medication-overuse headache vs control (P = .004) and migraine without aura subjects (P = .027); chronic migraine vs controlsubjects (P = .031).‡Medication-overuse headache vs control (P < .001) and migraine without aura subjects (P < .001); chronic migraine vs control(P = .005) and migraine without aura subjects (P = .037).§Medication-overuse headache and chronic migraine vs migraine without aura and control subjects (P < .001) (higher thanexpected prevalence of stress in both chronic headache groups in the adjusted residual analysis).¶Medication-overuse headache vs control subjects (P < .001).††Medication-overuse headache vs migraine without aura and control subjects (P < .001).‡‡Medication-overuse headache vs chronic migraine, migraine without aura and control subjects (P < .001).C = Caucasian.

4

1

2

3

4

56789

101112131415161718

19

20212223242526272829

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71


Scores on the Beck Depression Inventory andHamilton Anxiety Scale were higher in medication-overuse headache patients (mean � standard devia-tion: 14.65 � 11.21 and 23.24 � 11.01, respectively)compared to migraine without aura (5.47 � 4.00 fordepression, P = .027; 6.94 � 4.19 for anxiety, P < .001)and control subjects (3.29 � 3.05 for depression,P = .004; 4.29 � 3.93 for anxiety, P < .001). Chronicmigraine patients also had higher scores on theHamilton Anxiety Scale relative to migraine withoutaura (15.88 � 11.41 vs 6.94 � 4.19, P = .037) and con-trols subjects (15.88 � 11.41 vs 4.29 � 3.93, P = .005).In the Beck Depression Inventory, chronic migrainepatients exhibited higher scores than controls(8.65 � 6.50 vs 3.29 � 3.05, P = .031).

Medication-overuse headache and chronicmigraine patients were significantly more likely tohave higher percentages of stress (70.6% and 64.7%,respectively) than migraine without aura and controlsubjects (11.8% and 5.9%, respectively, P < .001). Themedication-overuse headache group also exhibitedhigher proportions of depression and anxiety than thecontrol group (52.9% vs 0% for depression, P < .001;58.8% vs 0% for anxiety, P < .001). The percentage ofanxiety in medication-overuse headache subjects wasclearly higher than in migraine without aura patients(58.8% vs 0%, P < .001).

The analysis for linearity used to test quantita-tive demographic and clinical variables in a gradu-ated manner (control, migraine without aura,chronic migraine, medication-overuse headache)revealed positive linearity for body mass index(from 23.5 � 3.25 in controls to 26.5 � 4.49 inmedication-overuse headache patients, P = .034) andscores on the Beck Depression Inventory (from3.29 � 3.05 to 14.65 � 11.21, P < .001) and HamiltonAnxiety Scale (from 4.29 � 3.93 to 23.24 � 11.01,P < .001).

Hematological Characteristics.—Medication-overuse headache patients had significantly highernumber of lymphocytes compared to migrainewithout aura patients during the ictal phase(2448.7/mm3 � 775.8 vs 1859.7/mm3 � 564.7, P = .027;Table 2). No other significant differences were foundin the hematological data between the groups orbetween the ictal and interictal phases of migrainewithout aura patients (Table 3). The aforementionedanalysis for linearity showed positive linearity forlymphocyte count (from 1961.7 � 385.6 in controlsto 2448.7 � 775.8 in medication-overuse headachepatients, P = .01) and hemoglobine (from 13.7 � 0.79to 14.6 � 1.31, P = .022), whereas negative linearityemerged for the percentage of CD8+ T lymphocytes(from 34.0 � 8.82 to 30.0 � 6.64, P = .046).

Table 2.—Hematological Characteristics of Patients and Control Subjects (Mean � SD)

Control SubjectsMigraine Without

Aura (Ictal) Chronic MigraineMedication-Overuse

Headache

Leucocytes/mm3 5747.0 � 1074.7 6882.3 � 2372.5 6664.7 � 1580.3 7029.4 � 1970.0Segmented leucocytes 3045.1 � 765.8 4229.4 � 2133.7 3802.4 � 1218.4 3777.4 � 1478.7Baston leucocytes 169.4 � 88.8 220.3 � 172.7 174.7 � 64.4 230.8 � 97.6Lymphocytes 1961.7 � 385.6 1859.7 � 564.7 2086.1 � 540.5 2448.7 � 775.8†Eosinophils 153.1 � 145.8 142.8 � 77.5 168.2 � 136.2 163.5 � 93.8Monocytes 393.0 � 105.1 416.8 � 134.1 402.5 � 134.6 394.5 � 107.0Basophils 21.4 � 31.2 13.1 � 24.8 18.4 � 34.8 11.0 � 24.5

Lymphocyte subpopulations (%)Natural killer cells 11.5 � 6.49 14.7 � 5.71 11.3 � 5.51 11.1 � 4.95B lymphocytes 11.0 � 3.06 10.1 � 5.55 10.7 � 3.76 8.5 � 2.90CD4 T lymphocytes 40.9 � 9.05 38.5 � 7.76 42.7 � 7.76 43.8 � 7.02CD8 T lymphocytes 34.0 � 8.82 32.6 � 7.06 28.7 � 5.67 30.0 � 6.64

Hemoglobin (g/dL) 13.7 � 0.79 14.2 � 1.04 14.2 � 1.07 14.6 � 1.31Hematocrit (%) 41.5 � 2.71 43.1 � 2.69 42.6 � 3.08 43.7 � 3.51

†Medication-overuse headache vs migraine without aura patients (P = .027).

Headache 5

1

2

34

5

6789

10111213141516171819

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76


The multiple linear regression analysis demon-strated that lymphocyte count remained associatedwith medication-overuse headache and migrainewithout aura groups (coefficient B = 540.7; confidenceinterval 95%: 55.2-1026.1; P = .03), but the percentageof lymphocyte variation attributed to this associationwas just moderate (R2 = 19.2%). There was a trend toa relation between body mass index and lymphocytecount (B = 37.3; CI 95%: -1.5-76.1; P = .059).

COMMENTSHematological data fell within the normal range

in all groups. However, lymphocyte count wassignificantly higher in medication-overuse headachepatients compared to migraine without aura subjectsin the headache phase. This finding can not beattributed to concomitant diseases or medicationsother than acetaminophen because these exclusioncriteria were screened in the recruitment of all sub-jects. Besides, there is no evidence that acetami-nophen produces increases in lymphocyte count oractivation at therapeutic doses.17 Immunologicaldisturbances were demonstrated only in the case ofacetaminophen-induced liver injury caused by toxicdoses,23 or in the presence of coexisting lymphoma.24

Overdose has been linked to genotoxic effects inhuman peripheral blood lymphocytes,25 and acetami-nophen may enhance the risk of developing leuke-mia.26 In any case, enrolled subjects neither surpassedthe daily limit dose of this analgesic, nor showed anyabnormality in routine blood liver function tests.

The literature contains myriad findings aboutimmune cells in episodic migraine.6 There are con-flicting data about the number of basophils,27,28 NKcells,29-33 B lymphocytes,30,32,33 T lymphocytes,29-34 andthe cytotoxic/suppressor T lymphocyte ratio.29,30,32

Discrepancies are most likely caused by the divergentpatterns of sample collection relative to the time ofthe attack.6 A lower CD8+ T lymphocyte count inmigraine patients compared to controls was alsoreported.33 Our study did not demonstrate any differ-ence in immune cell numbers between ictal and inter-ictal phases of migraine patients or between these andchronic migraine or control subjects.

Other findings regarding lymphocytes includelower b-adrenergic receptor sensitivity35 andenhanced dopamine D5-receptor expression.36

Recently, Nagata et al demonstrated that lympho-blasts derived from migraine patients exhibitedhigher serotonin levels relative to controls,37 as well asincreased expression of some genes, particularly thea-fodrin gene, a cytoskeletal protein abundantlyexpressed in the brain and linked to the corticalspreading depression, while the contrary wasobserved for other genes.38 Abnormal functioning ofNa/K ATPase, an enzyme linked to cortical spreadingdepression, in lymphocytes from patients withmigraine with aura was also reported.39 The triptans, apharmacological group commonly used for treatingmigraine, are able to decrease the cytotoxicity of NKcells and pro-matrix metalloproteinase-9 secretionfrom neutrophils.40

While episodic migraine has been investigatedwith regard to hematological data, this is not the casefor chronic forms of migraine. To our knowledgethere is a paucity of information concerning immunecells in chronic migraine, medication-overuse head-ache, chronic daily headache or even in the previouslydenominated transformed migraine. There is 1 reportof increased levels of interleukin (IL)-1b, IL-6 andtumor necrosis factor (TNF)-a in the supernatant of

Table 3.—Hematological Data From Migraine Without AuraPatients Collected During the Ictal and Interictal Phases

(Mean � SD)

Ictally Interictally

Leucocytes/mm3 6882.3 � 2372.5 6417.6 � 1454.8Segmented leucocytes 4229.4 � 2133.7 3628.8 � 1406.8Baston leucocytes 220.3 � 172.7 234.1 � 102.2Lymphocytes 1859.7 � 564.7 2024.0 � 535.1Eosinophils 142.8 � 77.5 126.5 � 106.8Monocytes 416.8 � 134.1 386.1 � 124.2Basophils 13.1 � 24.8 17.8 � 36.1

Lymphocytesubpopulation (%)Natural killer cells 14.7 � 5.71 14.5 � 4.92B lymphocytes 10.1 � 5.55 9.1 � 4.07CD4 T lymphocytes 38.5 � 7.76 39.7 � 7.58CD8 T lymphocytes 32.6 � 7.06 32.3 � 7.59

Hemoglobin (g/dL) 14.2 � 1.04 14.2 � 0.91Hematocrit (%) 43.1 � 2.69 42.4 � 2.57

No significant difference between groups.

6

123

4

5

6

789

101112131415161718192021

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95


peripheral blood mononuclear cell cultures from 4patients with chronic headache compared to con-trols.41 However, this publication does not completelyexplain the nature of such headaches.

Our results show that the lymphocyte countin peripheral blood was significantly higher inmedication-overuse headache patients than in ictalmigraine without aura subjects. Furthermore, posi-tive linearity for lymphocyte count along thespectrum control/migraine without aura/chronicmigraine/medication-overuse headache indicates thatthe “worse” the condition – assuming that chronicheadache with medication-overuse is a more severecondition than chronic or episodic migraines – thehigher the lymphocyte count. On the other hand,negative linearity was revealed for the percentage ofCD8+ T-lymphocytes. These findings remain to beexplained, but a few hypotheses may be formulated.

First, relative lymphocytosis may represent partof a chronic pro-inflammatory state closely linkedto the mechanisms underlying medication-overuseheadache. In this setting, the neurogenic inflamma-tion proposed for episodic migraine could be exacer-bated in chronic forms of migraine, especially inmedication-overuse headache, with systemic alter-ations detectable in peripheral blood. The paucityCD8+ T-lymphocytes is more difficult to explain, butmay be an adaptative response to neurogenic inflam-mation. Sarchielli et al described down-regulation ofleucocyte function antigen (LFA)-1 expression byCD4+ and CD8+ lymphocytes evaluated from theinternal jugular blood of migraine without aurapatients during attacks, parallel to transient increasein TNF-a, IL-6, and intercellular adhesion molecule(sICAM)-1.42 That structural alteration would be anattempt to antagonize the transvascular migration oflymphocytes, in a reaction to sterile inflammation inthe dura mater.42 In this setting, we hypothesize thatan exacerbated inflammatory state related to chronicheadache derived from episodic migraine could leadto more prominent changes in lymphocyte functionsand subsets. Other authors have found lower CD8+ Tlymphocyte counts in episodic migraine patients com-pared to controls,31,33 although this was not the case inour study. According to this 1st hypothesis, instead ofbeing caused by analgesic overuse, lymphocytosis

would represent an exacerbation of sterile neuro-genic inflammation that could predispose to analgesicoveruse.

Second, changes in lymphocyte counts and cellu-lar subsets may be a non-specific response to chronicpain. Other chronic painful conditions such ascomplex regional pain syndrome and fibromyalgiahave been associated with lower percentages ofCD8+ T lymphocytes compared to controls, althoughthe absolute numbers of lymphocytes were compa-rable.43 The decrease in CD8+ T lymphocytes mayresult from the stress associated with chronic pain,mediated through the activation of the sympatheticnervous system,43 as demonstrated in experimentalmodels of chronic stress.44,45 Chronic migrainepatients also had daily headache, but their lympho-cyte count were not different from controls or epi-sodic migraine subjects. This datum along with that ofother literature (referenced above) supports thestatement that the presence of chronic pain is not thecause of lymphocytosis in medication-overuse head-ache patients.

Third, a confounding factor common to both themedication-overuse headache group and the lym-phocyte count could have influenced the result. Toevaluate this hypothesis a multiple linear regressionanalysis was performed for variables that couldtheoretically be related to the variation in lympho-cyte number. Lymphocyte count remained associ-ated with medication-overuse headache andmigraine without aura groups, although these factorscould not explain more than 20% of the variation.Other tested variables did not reach statistical sig-nificance in this model, but body mass index showeda trend to positively influence lymphocyte count.Our study has limitations to perform an extensivemultiple linear regression analysis because of thesample size. Nevertheless, the attempt to adjust vari-ables with this model was valuable to evaluatepotential confounding factors. This may be particu-larly true for body mass index because obesity hasbeen linked to migraine chronification.11,12 High caf-feine intake is another risk factor for headachechronification, but was not assessed in patients andcontrol subjects. This represents a limitation of ourstudy.

Headache 7

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92


The finding of positive linearity for body massindex along the headache groups suggests that themore obese the patient, the higher the chance of suf-fering of chronic headache. Accordingly, Bigal et alshowed a positive association between body massindex and the disability caused by headache.46 Theputative mechanisms relating obesity to migraineare still under investigation, and the present body ofknowledge points to the development of a pro-inflammatory state in obese individuals. In experi-mental models, overweight and obesity have beenassociated to increased production of inflammatorymediators from adipose tissue T cells.47 Furthermore,adipocytokines such as adiponectin and leptin arerecognized as modulators of inflammation andare related to headache and cardiovascular events.Serum adiponectin levels and visceral obesity werefound to be increased in women with chronic dailyheadache over episodic migraine sufferers andhealthy controls.48 Although low serum leptin levelsand less fat percentage were demonstrated inpatients with migraine in comparison to control sub-jects,49 an association of hyperinsulinemia with highserum leptin levels in non-obese female migraineurswas shown.50 The aforementioned data along withthe positive linearity for body mass index demon-strated in our study rises some hypotheses: is therea similar variation of inflammatory modulators asadipocytokines among headache patients? Couldoverweight and obesity be responsible for pro-inflammatory changes that lead both to headachecronification and lymphocytosis? Positive linearityfor body mass index may also explain the samefinding for hemoglobine.

As there is a clear female preponderance inmigraine and chronic daily headache,1 the propor-tions of female and male subjects in the headachegroups of the present study are not unexpected.Gender matching was performed to avoid generatingdiscrepancies such as headache groups comprisedmostly by female patients and a control group with amajority of male subjects.

Depression and anxiety occur more often amongmigraineurs than in the general population.9 Thisco-morbidity may be important for non-migrainousheadache as well,9 and is one of the major risk factors

for transformation of episodic migraine into chronicmigraine or medication-overuse headache.10 Ourfindings endorse this association and suggest a gradu-ated worsening of psychopathological indices fromepisodic migraine to chronic forms of headache thatevolved from the former. However, multiple linearregression analysis did not attribute variation inlymphocyte number to the increase in depression andanxiety scores.To our knowledge, there is no report ofblood lymphocytosis associated with anxiety in theliterature, and only 1 study described higher lympho-cyte count in depressive adolescents over healthycontrols.51 On the other hand, alterations in the lym-phocyte subsets of patients with depression have beendisclosed, especially an increase of NK and regulatoryT cells.52,53

Our study has limitations to be considered: asmall sample size, the cross-sectional design, the sin-gular population composed of patients taking onlyacetaminophen, the limited parameters of immunesystem that were evaluated. Nevertheless, it raiseshypotheses to be assessed in future studies compris-ing more extensive investigation of immune profile inheterogeneous populations of headache patients.

In conclusion, our findings may indicate an exac-erbated chronic inflammatory state in patients withchronic headache derived from migraine, particularlyin medication-overuse headache. Besides, severalclinical and laboratorial characteristics may varyalong a spectrum extending from healthy subjects topatients suffering from chronic forms of migraine.

STATEMENT OF AUTHORSHIP

Category 1(a) Conception and Design

Cassiano M. Forcelini, Denise C. M. Dantas,Clarice Luz, Ricardo Santin, Airton T. Stein,Helena M.T. Barros, Liselotte M. Barea

(b) Acquisition of DataCassiano M. Forcelini, Ricardo Santin, ClariceLuz

(c) Analysis and Interpretation of DataCassiano M. Forcelini, Denise C. M. Dantas,Clarice Luz, Airton T. Stein, Helena M.T. Barros,Liselotte M. Barea

8

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

2 2


Category 2(a) Drafting the Manuscript


(b) Revising It for Intellectual ContentCassiano M. Forcelini, Denise C. M. Dantas,Clarice Luz, Airton T. Stein, Helena M.T. Barros,Liselotte M. Barea

Category 3(a) Final Approval of the Completed Manuscript


REFERENCES

1. Queiroz LP, Peres MF, Kowacs F, et al. Chronicdaily headache in Brazil: A nationwide population-based study. Cephalalgia. 2008;28:1264-1269.

2. Goadsby PJ. Migraine pathophysiology. Headache.2005;45(Suppl. 1):14-24.

3. Waeber C, Moskowitz MA. Migraine as an inflam-matory disorder. Neurology. 2005;64(Suppl. 2):9-15.

4. Davey G, Sedgwick P, Maier W, Visick G, StrachanDP, Anderson HR. Association between migraineand asthma: Matched case-control study. Br J GenPract. 2002;52:723-727.

5. Aamodt AH, Stovner LJ, Langhammer A, Hagen K,Zwart JA. Is headache related to asthma, hay fever,and chronic bronchitis? The Head-HUNT Study.Headache. 2007;47:204-212.

6. Kemper RHA, Meijler WJ, Korf J, Ter Horst GJ.Migraine and function of the immune system: Ameta-analysis of clinical literature publishedbetween 1966 and 1999. Cephalalgia. 2001;21:549-557.

7. Srikiatkhachorn A, Tarasub N, Govitrapong P.Effect of chronic analgesic exposure on the centralserotonin system:A possible mechanism of analgesicabuse headache. Headache. 2000;40:343-350.

8. Rainero I, Ferrero M, Rubino E, et al. Endocrinefunction is altered in chronic migraine patients withmedication-overuse. Headache. 2006;46:597-603.

9. Zwart JA, Dyb G, Hagen K, et al. Depression andanxiety disorders associated with headache fre-quency. The Nord-Trøndelag Health Study. Eur JNeurol. 2003;10:147-152.

10. Radat F, Creac’h C, Swendsen JD, et al. Psychiatriccomorbidity in the evolution from migraine to medi-cation overuse headache. Cephalalgia. 2005;25:519-522.

11. Dodick DW. Review of comorbidities and riskfactors for the development of migraine complica-tions (infarct and chronic migraine). Cephalalgia.2009;29(Suppl. 3):7-14.

12. Peterlin BL, Rapoport AM, Kurth T. Migraine andobesity: Epidemiology, mechanisms, and implica-tions. Headache. 2010;50:631-648.

13. Headache Classification Subcommittee of the Inter-national Headache Society. The international classi-fication of headache disorders, 2nd ed. Cephalalgia.2004;24(Suppl. 1):9-160.

14. Silberstein SD, Olesen J, Bousser M-G, et al onbehalf of the International Headache Society. TheInternational Classification of Headache Disorders,2nd ed (ICHD-II) – revision of criteria for 8.2Medication-overuse headache. Cephalalgia. 2005;25:460-465.

15. Headache Classification Committee: Olesen J,Bousser M-G, Diener H-C, et al. New appendix cri-teria open for a broader concept of chronicmigraine. Cephalalgia. 2006;26:742-746.

16. Gozzoli V, Treggiari MM, Kleger GR, et al. Ran-domized trial of the effect of antipyresis by metami-zol, propacetamol or external cooling onmetabolism, hemodynamics and inflammatoryresponse. Intensive Care Med. 2004;30:401-407.

17. Abramson SR, Weissmann G. The mechanisms ofaction of non-steroidal antiinflammatory drugs.Arthritis Rheum. 1989;32:1-9.

18. Beck AT, Ward CH, Mendelson M, Mock J, ErbaughG. An inventory for measuring depression. ArchGen Psychiatry. 1961;4:561-571.

19. Hamilton M. The assessment of anxiety states byrating. Br J Med Psychol. 1959;32:50-55.

20. Lipp M, Guevara A. Validação empírica doinventário de sintomas de stress. Estudos de Psico-logia. 1994;11:43-49.

21. Beck AT, Steer RA, Garbin MG. Psychometricproperties of the beck depression inventory:Twenty-five years of evaluation. Clin Psychol Rev. 1988;8:77-100.

22. Hayley S, Poulter MO, Merali Z, Anisman H. Thepathogenesis of clinical depression: Stressor- andcytokine-induced alterations of neuroplasticity.Neuroscience. 2005;135:659-678.

Headache 9

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96


23. Masson MJ, Peterson RA, Chung CJ, Graf ML,Carpenter LD, Ambroso JL. Lymphocyte lossand immunosupression following acetaminophen-induced hepatotoxicity in mice as a potential mecha-nism of tolerance. Chem Res Toxicol. 2007;20:20-26.

24. Inoue D, Kimura T, Shimoji S, et al. Angioiimuno-blastic T-cell lymphoma complicated by recurrentdrug-induced agranulocytosis. Rinsho Ketsueki.2009;50:87-91.

25. Ibrulj S, Rahmanovic A, Haveric S, Haveric A, PasicAD. Cytogenetic evaluation of paracetamol effectsin human lymphocytes culture. Drug Chem Toxicol.2007;30:133-143.

26. Weiss JR, Baker JA, Baer MR, Menezes JR, NowellS, Moysich KB. Opposing effects of aspirin andacetaminophen use on risk of adult acute leukemia.Leuk Res. 2006;30:164-169.

27. Heatley RV, Denburg JA, Bayer N, Bienenstock J.Increased plasma histamine levels in migrainepatients. Clin Allergy. 1982;12:145-149.

28. Thonnard-Neumann E, Neckers LM. T-lymphocytesin migraine. Ann Allergy. 1981;47:325-327.

29. Covelli V, Maffione AB, Munno I, Jirillo E. Alter-ations of nonspecific immunity in patients withcommon migraine. J Clin Lab Anal. 1990;4:9-15.

30. Leone M, Sacerdote P, D’Amico D, Panerai AE,Bussone G. Beta-endorphin concentrations in theperipheral blood mononuclear cells of migraine andtension-type headache patients. Cephalalgia. 1992;12:154-157.

31. Gilman-Sachs A, Robbins L, Baum L. Flow cyto-metric analysis of lymphocyte subsets in peripheralblood of chronic headache patients. Headache.1989;29:290-294.

32. Martelletti P, Sutherland J, Anastasi E, Di Mario U,Giacovazzo M. Evidence for an immune-mediatedmechanism in food-induced migraine from a studyon activated T-cells, IgG4 subclass, anti-IgG anti-bodies and circulating immune complexes. Head-ache. 1989;29:664-670.

33. Leone M, Biffi M, Leoni F, Bussone G. Leucocytesubsets and cortisol serum levels in patients withmigraine without aura and chronic tension-typeheadache. Cephalalgia. 1994;14:139-142.

34. Massari F, D’Andrea L, Cervo MA, Serra FP,Covelli V, Buscaino GA. Quantitative and qualita-tive modifications of lymphocyte subsets after sub-lingual administration of isosorbide dinitrate inmigraineurs. Preliminary report. Acta Neurol(Napoli). 1994;16:11-18.

35. Kuritzky A, Bennet E, Hering R, Ebstein R.Reduced sensitivity of lymphocyte beta-adrenergicreceptors in migraine. Headache. 1993;33:198-200.

36. Barbanti P, Bronzetti E, Ricci A, et al. Increaseddensity of dopamine D5 receptor in peripheralblood lymphocytes of migraineurs: A marker formigraine? Neurosci Lett. 1996;207:73-76.

37. Nagata E, Hamada J, Shimizu T, et al. Altered levelsof serotonin in lymphoblasts derived from migrainepatients. Neurosci Res. 2007;57:179-183.

38. Nagata E, Hattori H, Kato M, et al. Identification ofbiomarkers associated with migraine with aura.Neurosci Res. 2009;64:104-110.

39. Scarrone S, Podestà M, Cupello A, et al. Abnormali-ties of Na/K ATPase in migraine with aura. Cephala-lgia. 2007;27:128-132.

40. Puente J, Jaque M, Carrasco C, et al. Triptan drugs,natural killer cell cytotoxicity, and neutrophils pro-matrix metalloproteinase-9 secretion. Headache.2008;48:1482-1489.

41. Jeong HJ, Hong SH, Nam YC, et al. The effectof acupuncture on proinflammatory cytokineproduction in patients with chronic headache: Apreliminary report. Am J Chin Med. 2003;31:945-954.

42. Sarchielli P, Alberti A, Baldi A, et al. Proinflamma-tory cytokines, adhesion molecules, and lymphocyteintegrin expression in the internal jugular blood ofmigraine patients without aura assessed ictally.Headache. 2006;46:200-207.

43. Kaufmann I, Eisner C, Richter P, et al. Lymphocytesubsets and the role of TH1/TH2 balance in stressedchronic pain patients. Neuroimmunomodulation.2007;14:272-280.

44. Kubera M, Basta-Kaim A, Holan V, et al. Effect ofmild chronic stress, as a model of depression, on theimmunoreactivity of C57BL/6 mice. Int J Immunop-harmacol. 1998;20:781-789.

45. Engler H, Dawils L, Hoves S, et al. Effects of socialstress on blood leucocyte distribution: The role ofalpha- and beta-adrenergic mechanisms. J Neuro-immunol. 2004;156:153-162.

46. Bigal ME, Tsang A, Loder E, Serrano D, Reed ML,Lipton RB. Body mass index and episodic head-aches. Arch Intern Med. 2007;167:1964-1970.

47. Yang H, Youm YH, Vandanmagsar B, et al. Obesityincreases the production of proinflammatory media-tors from adipose tissue T cells and compromisesTCR repertoire diversity: Implications for systemic

10

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100


inflammation and insulin resistance. J Immunol.2010;185:1836-1845.

48. Peterlin BL, Alexander G, Tabby D, ReichenbergerE. Oligomerization state-dependent elevations ofadiponectin in chronic daily headache. Neurology.2008;70:1905-1911.

49. Guldiken B, Guldiken S, Demir M, Turgut N, TugrulA. Low leptin levels in migraine: A case controlstudy. Headache. 2008;48:1103-1107.

50. Bernecker C, Pailer S, Kieslinger P, et al. GLP-2 andleptin re associated with hyperinsulinemia in non-obese female migraineurs. Cephalalgia. 2010;30:1366-1374.

51. Schleifer SJ, Barlett JA, Keller SE, Eckholdt HM,Schiflett SC, Delanev BR. Immunity in adolescentswith major depression. J Am Acad Child AdolescPsychiatry. 2002;41:1054-1060.

52. Ravindran AV, Griffiths J, Merali Z, Anisman H.Circulating lymphocyte subsets in major depressionand dysthymia with typical or atypical features. Psy-chosom Med. 1998;60:283-289.

53. Li Y, Xiao B, Qiu W, et al. Altered expression ofCD4(+)CD25(+) regulatory t cells and its 5-HT(1a)receptor in patients with major depression disorder.J Affect Disord. 2010;124:68-75.

Headache 11

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25


AUTHOR QUERY FORM

Dear Author,During the preparation of your manuscript for publication, the questions listed below have arisen. Please

attend to these matters and return this form with your proof.Many thanks for your assistance.

QueryReferences

Query Remark

q1 AUTHOR: All the text marked with strikethrough in the original paper havebeen deleted. Please confirm that this is correct.

q2 AUTHOR: Please check whether the spelling of “cronification” is correct?

Toppan Best-set Premedia LimitedJournal Code: HEAD Proofreader: EmilyArticle No: 1902 Delivery date: 27 April 2011Page Extent: 11 Copyeditor: Ivan

Analysis of Leukocytes in Medication-Overuse Headache, Chronic Migraine, and Episodic Migraine

Documents

Transcript of Analysis of Leukocytes in Medication-Overuse Headache, Chronic Migraine, and Episodic Migraine