SEARCH

SEARCH BY CITATION

Keywords:

  • migraine;
  • chronic migraine;
  • post-traumatic stress disorder;
  • abuse

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES

Objective.— To assess and contrast the relative frequency of self-reported post-traumatic stress disorder (PTSD) in patients with episodic migraine and chronic/ transformed migraine.

Background.— Several risk factors have been identified as risk factors for chronification of headache disorders. Childhood abuse has been suggested as a risk factor for chronic pain in adulthood. In addition depression, as well as several other psychiatric disorders, are co-morbid with migraine. Recent data suggest that PTSD may be more common in headache sufferers than in the general population.

Methods.— This was a prospective, pilot study conducted at a headache center. Adult subjects with episodic, chronic, or transformed migraine were included. Demographic information, depression history, body mass index (BMI), and headache characteristics were obtained. PTSD was assessed using the life events checklist (LEC) and the PTSD checklist, civilian version (PCL-C). We contrasted the data from episodicmigraineurs and chronic/transformed migraine participants (CM) and conducted multivariate analyses, adjusting for covariates.

Results.— Of the 60 participants included, 91.7% were female with a mean age of 41.4 ± 12.5 years old. EM was diagnosed in 53.3% and CM in 46.7%. The mean BMI was not significantly different between groups. In contrast, the relative frequency of depression was significantly greater in subjects with CM (55.2%) than EM (21.9%, P = .016). There was no significant difference in the percentage of participants reporting at least 1 significant traumatic life event (LE) or in the mean number of traumatic LEs between EM and CM participants. However, the relative frequency of PTSD reported on the PCL in CM (42.9%) was significantly greater as compared to EM (9.4%, P = .0059. After adjusting for depression and other potential confounders, the difference remained significant P = .023).

Conclusion.— PTSD is more common in CM than in episodic migraineurs. This suggests that PTSD may be a risk factor for headache chronification, pending longitudinal studies to test this hypothesis.

Post-traumatic stress disorder (PTSD) occurs as a result of exposure to extreme traumatic stressors that arouse feelings of intense fear, helplessness, and horror in exposed individuals. As a result of these stressors the individual's response characteristically involves emotionally re-experiencing the event, numbing of affect, and avoidance of stimuli, which are associated with the event, as well as increased arousal.1,2

Both migraine and PTSD are more prevalent in women than men.3 For migraine, hormonal influences, at least in part, explain the female predominance.4-6 For PTSD, it has been suggested that the higher prevalence may be due to the higher rates of physical and sexual abuse in women.3

The National Violence Against Women Survey reported that 1 in 6women(16.67%) in the United States has experienced an attempted or completed rape at some time in their lives and that 26% of women are victimized by intimate partners in the form of physical assault, rape, or stalking during the course of a lifetime.7 Given the relatively high prevalence of abuse in the general population, it is not surprising that recent studies have also suggested an association between abuse and headache.8,9 In addition, de Leeuw et al have suggested that headache sufferers may have an increased risk of PTSD as compared to the historically reported prevalence for the general population.10

Accordingly, our study was designed to compare and contrast the relative frequency of self-reported PTSD, using a validated PTSD survey, in persons with episodic migraine (EM) as compared to chronic and transformed migraine. Although the definitions of chronic migraine (CM) and transformed migraine are different, as both refer to the same disease status (the result of migraine chronification), for clarity purposes we refer to this group as CM. Thus, we hypothesized that chronic migraineurs would report a greater relative frequency of PTSD than episodic migraineurs.

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES

Consecutive headache patients from the ages of 18-65 years old presenting for evaluation to an outpatient headache center, from September 2006 to December 2006, were asked to complete a written survey. Subjects who fulfilled criteria for EM, CM, or transformed migraine were included in the study. EM and CM were diagnosed according to the criteria defined by the second edition of the international classification of headache disorders (ICHD-2);11 transformed migraine was diagnosed according to the criteria proposed by Silberstein and Lipton.12 Herein, the group of participants withCMand transformed migraine will be collectively referred to as CM.

After agreeing to participate, subjects completed a semi-standardized questionnaire constructed for this study. This questionnaire ascertained self-reported demographic information (including age, gender, race, marital status, education, and total household income), history of a medical diagnosis of depression, self-reported height and weight, headache diagnoses and headache characteristics including location, quality of pain, duration, frequency, and migraine-associated symptoms. In addition, subjects completed the life events checklist (LEC) and the PTSD checklist, civilian version (PCL-C).13

MEASURES DERIVED FROM THE QUESTIONNAIRE

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES

Body Mass Index (BMI).— As recent studies have shown an association between obesity and migraine, as well as an association with obesity and depression in migraine,14-16 BMI was calculated for all participants. The following formula was used: BMI = wt (lbs)/ht2 (× 703 and 6 categories were defined as follows: underweight (<18.5), normal weight (18.5-24.9), overweight (25-29.9), obesity grade I (30-34.9), obesity grade II (35-39.9), and obesity grade III (≥40).

Life Event Check-list (LEC).— The LEC is a checklist used to screen for events that meet PTSD criteria for DSM-IV Criterion A.1 It includes 17 questions regarding potentially difficult or stressful LEs that sometimes happen to people. Participants were asked to report if the event or events(s) happened to them personally, they witnessed it, learned about it, were not sure or it did not apply. The events included natural disasters, fire or explosion, serious accidents at work, home or during recreational activity, physical assault, assault with a weapon, sexual assault, combat, captivity, life-threatening illness or injury, severe human suffering, sudden, violent death, unexpected death of someone close to participant, serious injury, and harm or death the participant caused to someone else. The total number of traumatic LEs in each category was assessed for each individual category (“happened to me,”“witnessed it,” and “learned about it”), as well as with all 3 categories combined for all study participants and again for all participants who fulfilled PTSD criteria based on the PCL.

Post-Traumatic Stress Disorder Check List – Civilian Version (PCL-C).— The PCL-C is a 17 question, self-report measure used to assess PTSD criterion B-D from the DSM-IV criteria.1 For each question, the participant is asked to indicate how much they have been bothered by each symptom on a 5-point Likert scale from 1 (not at all) to 5 (extremely) in regards to their most significant life event (LE) stressor. The PCL-C has exhibited test – retest stability, an overall internal consistency (α = 0.94) and provides a valid and reliable assessment of the presence of PTSD, with a sensitivity of 0.944 and specificity of 0.864 as compared to the clinician-administered PTSD scale. A total score of ≥44 is considered clinically significant for PTSD.13

ANALYSIS

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES

Three groups were defined based on the category of reported traumatic LEs. Group 1 had at least 1 traumatic LE on the LEC in the “happened to me” category, with a score of ≥44 on the PCL-C. Group 2 had at least 1 traumatic LE on the LEC in the “happened to me” or “witnessed it” categories, with a score of ≥44 on the PCL-C. Group 3 had at least 1 traumatic LE on the LEC in the “happened to me,”“witnessed it,” or “learned about it” categories, with a score of ≥44 on the PCL-C. All statistical analyses were performed using Data Desk Version 6 by Data Description (Ithaca, NY, USA). Differences in continuous variables were tested using ANOVA. Fisher's exact test was used for categorical variable comparisons. Logistic regression was used for models adjusting for demographics and control variables including age, sex, marital status, mean income, BMI, and depression. The study and all questionnaires were approved by the Investigational Review Board.

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES

Demographics.— Of the 74 patients who completed the survey, 12 were excluded as they had diagnoses other than migraine or CM, and 2 were <18 years old. The majority of participants were Caucasian females. The mean (±SD) age of all participants was 41.4 ± 12.5. Of participants, 53.3% (32/60) had EM and 46.7% (28/60) had CM. There were no significant differences between the CM and EM in respect to age, sex, race, marital status, mean income, or BMI (Table 1).

Table 1.—. Demographics of Participants
 Episodic MigraineChronic MigraineP Value
N (%)32 (53.3)32 (46.7) 
Mean Age43.3 ± 13.639.2 ± 10.8.21
Female (%)90.692.91.00
Caucasian (%)81.396.4.11
Marital Status  .87
 Single (%) 37.542.9 
 Married (%) 43.842.9 
 Divorced/widowed (%)18.814.3 
Mean Income  .55
 <$20,000 (%)12.921.4 
 $20-50,000 (%) 38.725.0 
 $50-100,000 (%) 25.842.9 
 >$100,000 (%)22.610.7 
Hx Depression (%)21.955.2.016
Mean BMI30.1 ± 21.927.7 ± 7.9.59

Depression.— Aprior medical history of depression was reported in 22 of the 60 total participants (36.7%) and it was reported in a significantly greater number of CM participants. While 21.9% (7/32) of the mi-graineurs reported a prior medical diagnosis of depression, it was reported in 53.6% (15/28) of CM participants (P = .016) (Table 1).

LEC.— There was no significant difference in the percentage of participants with EM as compared to CM participants reporting at least 1 significant traumatic life event (LE) as “happened to me,”“witnessed it,” or “learned about it” individually or when the groups were combined. Of the total participants, 78.1% (25/32) of EM and 78.6% (22/28) reported at least 1 significant traumaticLEthat happened to them, irrespective of whether or not they fulfilled PTSD criteria (no significance) (Table 2).

Table 2.—. Traumatic Life Events Reported on the Life Event Checklist (LEC), Irrespective of Whether or Not Participants Fulfilled PTSD Criteria
Category of Life Event (LE) ReportedMigraine (n = 32)CM (n = 28)P Value
≥1 LE in the “happened to me” category (%)78.178.61.0
≥1 LE in the “witnessed it” category (%)59.467.9.59
≥1 LE in the “learned about it” category (%)68.864.3.79
≥1 LE in the “happened to me” or “witnessed it” categories (%)96.989.3.33
≥1 LE in the “happened to me,”“witnessed it,” or “learned about it” categories (%)10092.9.21
Mean number of traumatic life events reported on the LEC irrespective of fulfillment of PTSD criteria
“Happened to me”2.47 ± 2.412.71 ± 22.42.70
“Witnessed it”1.06 ± 21.341.5 ± 21.5.24
“Learned about it”2.5 ± 23.142.57 ± 22.99.93
“Happened to me” or “witnessed it”3.53 ± 22.964.21 ± 23.13.39
“Happened to me,”“witnessed it,” or “learned about it”6.03 ± 25.126.79 ± 25.07.57

Finally we assessed if there was a sequential relationship between the numbers of prior traumatic LEs reported as “happened to me” by EM and CM participants who fulfilled PTSD criteria on the PCL. It is interesting to note that as the number of traumatic LE increased, the CM/EM index increased, up until 3 LEs. However, with each increase in the number of traumatic LEs, the number of participants being analyzed also increasingly declined; and after 3 LEs (and most likely due to the shrinking sample size) the relationship stabilized (Table 3).

Table 3.—. The CM/EM Index of Increasing Numbers of Traumatic Life Events Reported as “Happened To Me” in Participants Fulfilling PTSD Criteria on the PCL
# of Traumatic Life Events Reported as “Happened to Me”EM (n = 32)CM (n = 28)CM/EM IndexP ValueP Value*
  • *

    Corrected P value for confounders including depression.

≥19.4%42.9%4.6.0059.0230
≥26.3%39.3%6.5.0035.032
≥33.1%28.6%9.2.0091.085
≥43.1%21.4%6.9.0430.13
≥63.1%14.3%4.6.17.21

PTSD.— In both groups 2 and 3, 4/32 (12.5%) EM participants fulfilled criteria for PTSD on the PCL, in contrast to 12/28 (42.9%) of CM participants, P = .01. In the most stringently defined group, group 1, the relative frequency of self-reported PTSD remained significantly greater in CM than EM, with 3/32 (9.5%) of EM and 12/28 (42.9%) of the CM participants fulfilling PTSD criteria on the PCL, P = .0059. This remained significant after adjusting for age, sex, marital status, income, BMI, and depression, P = .023, (Fig. 1).

image

Figure 1.—. Of CM participants, 42.9% fulfilled criteria for PTSD on the PCL as compared to 9.4% of EM, both when uncorrected for confounders (P = .0059) and when corrected for confounders (P = .023).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES

In our study, the number of traumatic LEs that were self-reported to have directly happened to our participants was 78%, while 93% reported traumatic LEs as having directly happened to them or that they witnessed. These findings are comparable to what was found in the 1996 Detroit area survey by Breslau et al. In Breslau's study the lifetime prevalence of exposure to one or more traumatic events was 89.6%.2

In addition, we found the relative frequency of self-reported PTSD to be significantly greater in CM participants as compared to EM, with a relative frequency of 43% and 9%, respectively. And although our study lacked non-headache controls, and direct comparisons cannot truly be made with the general population, our findings in CM do appear to be greater than the prevalence of PTSD, which has been reported in the general population of approximately 9-14%.2,3

In a previous study evaluating the relative frequency of self-reported PTSD in pain subjects, de Leeuw et al evaluated a group of headache patients (which combined tension-type headache subjects and migraineurs) vs masticatory muscle pain subjects. The headache frequency was not reported in their study. However, a PTSD relative frequency of 16% was found in their headache group, which is comparable to our finding of 9.4% in EM.10

In addition, the relative frequency of PTSD reported on the PCL in CM participants in this study is also comparable to the relative frequency of what Tietjen et al found in regards to sexual abuse in migraineurs, 30%, and similar, to what Peterlin et al found in regards to a history of physical and/or sexual abuse in CDH (40%).8,9

Although our data indicate a higher frequency of self-reported PTSD amongCMparticipants than EM, cautions are required when interpreting our results. First, PTSD criteria were assessed using the PCL survey; and although it is a validated survey shown to have excellent sensitivity and specificity as compared to the clinician administered PTSD Scale it is not by a physician. Second, our study lacked a general population control and thus, no direct comparison can be made in regards to PTSD frequency between migraineurs and the general population. Next, depression is one of the most common comorbid features of PTSD in both male and female trauma survivors.17 It is also a well-recognized co-morbid condition with migraine.18 And, although our current study inquired about a physician-diagnosis of depression, a validated questionnaire to capture all subjects with depression may have captured more accurately and completely all subjects with depression than our current study. Our study also made no attempt to delineate at what age the traumatic life stressors occurred. This may be important as a meta-analysis by Davis et al has shown evidence that individuals who report abusive or neglectful childhood experiences are at an increased risk of chronic pain as adults.19 Finally, our sample size was small. Further large-scale studies are warranted to more definitively evaluate the findings of this preliminary study.

CONCLUSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES

Migraine and PTSD are relatively common presenting problems. Our pilot study suggests that PTSD occurs with greater frequency in patients with CM than EM. Thus PTSD may be a risk factor for headache chronification, pending longitudinal studies to test this hypothesis. As a previous study has suggested that PTSD treatment alone can positively influence chronic pain conditions and disability,20 one of the implications of this study is that there should be greater consideration for the evaluation of PTSD in CM patients, as well as for the use of cognitive/behavioral therapy (alone or in combination with pharmacological therapy) in this subgroup of headache sufferers.

Acknowledgments

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES

Acknowledgments: Special thanks and appreciation to Dr. Ann Scher for her helpful comments and suggestions.

REFERENCES

  1. Top of page
  2. Abstract
  3. METHODS
  4. MEASURES DERIVED FROM THE QUESTIONNAIRE
  5. ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. CONCLUSION
  9. Acknowledgments
  10. REFERENCES