New Daily Persistent Headache in Children and Adolescents
- First Online:
- Cite this article as:
- Baron, E.P. & Rothner, A.D. Curr Neurol Neurosci Rep (2010) 10: 127. doi:10.1007/s11910-010-0097-3
- 211 Views
New daily persistent headache (NDPH) is a form of chronic daily headache (CDH) that may have features of both migraine and tension-type headache. In contrast with other types of CDH, NDPH is characterized by patients recalling the specific date their unremitting daily headache began. In comparison, chronic tension-type headache and chronic migraine are preceded by a gradually increasing frequency of headache. After several months, all three of these CDH forms often have a similar phenotype, making early history a key to diagnosing NDPH. Evaluations to exclude secondary causes are necessary but usually negative. NDPH is difficult to treat and requires a multimodal approach. Questions regarding NDPH remain unanswered. Additional prospective studies are necessary to further understand, characterize, diagnose, and treat NDPH.
KeywordsNew daily persistent headacheChronic daily headacheChildrenAdolescentsChronic migraineChronic tension-type headache
New daily persistent headache (NDPH), first described in 1986 , is one of four types of chronic daily headache (CDH); the others are transformed (chronic) migraine (CM), chronic tension-type headache (CTTH), and hemicrania continua (HC) . CDH is defined as a headache lasting more than 4 h/day on more than 15 days/month for more than 3 months. According to the International Classification of Headache Disorders–II , NDPH is a headache that is present daily within 3 days of onset and is unremitting for more than 3 months. It must have two of the following four pain characteristics: bilateral location, pressing/tightening (nonpulsating) quality, mild or moderate intensity, and not aggravated by routine physical activity such as walking. It must have both of the following: no more than one of photophobia, phonophobia, or mild nausea, and neither moderate or severe nausea nor vomiting. It cannot be attributed to any other disorder.
In a study of mostly adult NDPH patients, 82% remembered the exact date their headache began , whereas a study of children and adolescents with NDPH reported 48.1% could recall the month their headaches began [4•]. Additionally, about 30% of patients with NDPH have a family history of headache, usually migraine. In our experience, there often is a history of occasional mild headaches or occasional migraine before the onset of NDPH.
Comorbid symptoms in pediatric NDPH patients include sleep disturbance in at least two thirds (especially delay in falling asleep), dizziness, generalized “weakness,” visual disturbances such as blurred vision, abdominal pain, occasional vomiting, sore neck muscles, difficulty concentrating, fatigue, and other nonspecific symptoms [6, 7]. Mood disorders such as anxiety and depression sometimes are associated with NDPH, although these often follow months of persistent symptoms rather than being a preexisting or early symptom . Frequent school absences and overuse of over-the-counter medications commonly are seen. In one study, nine of 34 pediatric patients (26%) with NDPH had excessive school absences (>15 absences/year due to headache) and nine of 34 (26%) overused analgesics three or more times weekly for longer than 6 months . In addition, we find that many patients are straight-A students and are overcommitted to multiple activities outside school. They experience higher stress levels related to maintaining their good grades and extracurricular activities. Some patients coexist with their daily headache and do not overuse medications, miss school, have a drop in grades, or have a decrease in extracurricular activities. Others frequently overuse medications, miss school, drop their grades, decrease their extracurricular activities, and are more severely affected. The explanation behind these clinical differences is unknown and requires further research.
In general population studies, CDH is estimated to affect 4% to 5% of adults [8, 9] and 1% of adolescents . Prevalence rates have been reported as 2.4% in girls and 0.8% in boys . Prevalence rates of NDPH in the general population have been reported at 0.03%  to 0.1% .
In tertiary headache centers, CDH is seen in 30% of patients . Of these patients, NDPH is diagnosed more frequently in children and adolescents than in adults . NDPH has been estimated to occur in 1.7% to 10.8% of adults with CDH [14, 15], and in 13% to 35% of children with CDH [16–18]. It is more common in females, with the female/male ratio in adults at 2.5:1  and in children and adolescents at 1.8:1 [4•]. It has an earlier age of onset in women (ages 16–35) than in men (ages 26–45) . One study of 680 pediatric headache clinic patients reported 189 (28%) with tension-type headache, 34 of whom met NDPH criteria . This represented 5% of the entire headache population studied and 18% of the CDH patients, consistent with other prevalence studies. Of these NDPH patients, 24 were female and ten male, consistent with prior studies.
The pathophysiology of NDPH is unknown, although some general associations have been made with infection, extracranial surgery, stressful life events, hypothyroidism, hypertension, analgesic overuse, and frequent caffeine use [5, 19–21]. In one study of 34 pediatric headache patients with NDPH, five (15%) had a preceding infection, five (15%) had a preceding mild to moderate head injury, and four had significant psychosocial issues . Another study reported that in 23% of patients, NDPH was preceded by mild head trauma . Overall, infections seem to be most frequently associated with NDPH. One study reported 27 of 32 NDPH patients (84%) with evidence of active Epstein-Barr virus infection . An episode of influenza or a “cold” preceded NDPH onset in 30% of patients in one study  and 43% in another . Other infectious associations have been made with herpes simplex virus, cytomegalovirus, Salmonella, adenovirus, Toxoplasma, and herpes zoster [1, 20, 24]. Given the variety of infectious associations, it has been postulated that the nonspecific inflammatory response initiated by the infection, rather than the infectious agent itself, may be the trigger . In children, the most frequent triggers for NDPH have been reported to include febrile illness, minor head trauma, and extracranial surgery . No inciting factor was identified in 12% of NDPH patients in one study , although a Japanese study of NDPH reported that precipitating events could not be identified in 65% of patients . The latter is more consistent with our clinical practice.
History and examination are the most critical aspects in evaluating NDPH, particularly when the patient reports a specific date the headache began. The recollection of a specific date of onset is characteristic and the most useful clue to diagnosis. The other three types of CDH (CTTH, CM, and HC) should be ruled out, and history alone should differentiate these. Although HC is included in this CDH group, some authorities feel it should be excluded. HC presents much differently and usually can be separated easily from the other three types of CDH given its characteristic unilateral location, autonomic features, and absolute response to indomethacin. For these reasons, we exclude it from our discussion of CDH. The remaining three CDH types (NDPH, CTTH, and CM) all have variably overlapping symptoms, making the history of abrupt onset the most useful diagnostic clue. After several months of symptoms, all three of these types often have the same phenotype, suggesting there are several ways to get to CDH, but once there, the phenotype is the same.
Patients with NDPH should have no symptoms of increased intracranial pressure or other neurologic symptoms. Their symptoms should not be progressive. They should have a normal general and neurologic examination and no papilledema. Any secondary headache may come on suddenly, so it is imperative to rule out secondary causes of new-onset daily headache. Secondary causes of NDPH include cerebral venous sinus thrombosis, low or high cerebrospinal fluid (CSF) pressure, carotid or vertebral dissection, vasculitis or other vascular disorders, toxins, sphenoid sinusitis, temporomandibular joint disease, congenital abnormalities (eg, symptomatic Chiari malformation), metabolic disorders, endocrine disorders (eg, Hashimoto’s thyroiditis), hormonal changes, neoplasm, infection (particularly viral), and posttraumatic headache .
Routine blood work should be checked and should include thyroid studies, sedimentation rates, complete blood count, and complete metabolic profile. Additional testing should be tailored toward information discovered during history taking and examination, such as symptoms of autoimmune processes and specific infections (eg, Lyme disease).
Diagnostic studies, although frequently pursued, have a low diagnostic yield in the setting of a negative history and neurologic examination. Although electroencephalography may be considered if there is concern about seizures or a focal lesion, it is not routinely indicated.
According to the American Academy of Neurology practice parameters, neuroimaging should not be done routinely in children with recurrent headaches and normal neurologic examinations . The variables predicting space-occupying lesions include headache of less than 1 month’s duration, absence of family history of migraine, abnormal neurologic examination, gait abnormalities, and seizures . Despite these recommendations, neuroimaging is performed frequently, usually because of parental anxiety and medicolegal concerns. In general, an abrupt onset of daily headache warrants neuroimaging, and an MRI of the brain is preferable to CT. Adding magnetic resonance angiography (MRA) and magnetic resonance venography (MRV) is a reasonable consideration based on the clinical presentation. Overall, neuroimaging likely will be normal, unless there is a history suggesting a secondary cause, or focal neurologic symptoms or signs are present.
In one study, neuroimaging with CT or MRI was performed in 30 children with CDH and a normal neurologic examination, and five children (16.6%) had abnormal findings . These findings included arachnoid cysts, Chiari I malformations, sinus disease, occult vascular malformations, and dilated Virchow-Robin spaces, all of which were considered incidental and unrelated to the headache . In another study, 49 patients with NDPH underwent MRI or CT . Of these, 66% had normal results, whereas the remainder had nonspecific and unrelated findings . In another study, 241 of 681 pediatric headache clinic patients underwent MRI (218) or CT (23) of the brain . Of these, 46 (19%) had benign abnormalities that did not change management, including sinus disease, Chiari I malformations, nonspecific white matter abnormalities, venous angiomas, arachnoid cysts, enlarged Virchow-Robin spaces, pineal cysts, mega cisterna magna, fenestration of the proximal basilar artery, and periventricular leukomalacia. In addition, 23 patients (9.5%) had findings requiring a change in management, including sinus disease, tumors, old infarcts, Chiari I malformations, moyamoya disease, intracranial vascular stenosis, internal jugular vein occlusion, arteriovenous malformation, demyelinating disease, and intracerebral hemorrhage. Ultimately, this study concluded that approximately 20% of pediatric headache patients with neuroimaging have benign imaging findings that do not result in a change in headache management. Furthermore, imaging findings requiring a change in management are rare in the absence of abnormal neurologic symptoms and signs, occurring in only 1.2% of patients.
Lumbar puncture is not indicated unless the history suggests a possible infectious, inflammatory, or pressure-related etiology. Caution is advised in obese and very anxious patients because falsely elevated opening pressures may be seen, resulting in a misdiagnosis of idiopathic intracranial hypertension (pseudotumor cerebri). Sedation and fluoroscopy should be considered in this setting to avoid falsely elevated pressures, or as we refer to it, pseudo-pseudotumor.
A 15-year-old girl presented with her parents for evaluation of daily headaches over a 1-year period. Both the patient and her parents were very anxious. She reported that her headaches began on November 19 of the previous year. Neither she nor her parents could recall any changes in her routine or other obvious reasons to explain her new headache. They did recall that she left school early the day before the headache began because of mild flulike symptoms, which resolved quickly. The patient was otherwise healthy but had rare “normal,” mild headaches during periods in which she was stressed with schoolwork. She was a straight-A student and involved in multiple after-school activities. Her mother had episodic migraine.
The patient described her headache as a dull, pressing sensation throughout her head. The pain fluctuated between mild and moderate but never became severe. Despite this fluctuation, she admitted that she had not had a headache-free day since the headaches began, and there was some level of discomfort 24 hours a day, 7 days a week. The patient had occasional photophobia when her headache became stronger but no phonophobia, nausea, or vomiting. She felt that her concentration was impaired.
The patient’s vital signs and general examination were normal, except for some mild tenderness over the cervical paraspinal musculature. Detailed neurologic and funduscopic examinations were normal. She recently had had an extensive workup, which was unremarkable. Routine blood work included thyroid studies, inflammatory markers, complete blood count, and metabolic profile. Neuroimaging included MRI, MRA, and MRV of the brain, which demonstrated a small arachnoid cyst. One physician stated that the headaches may be the result of this cyst and would have to be followed up with serial MRI. The patient and her parents were very anxious about this cyst as the cause of her headaches.
The patient had trials of topiramate and gabapentin and had been using over-the-counter medications most days of the week for the previous 6 months. Her school absences had increased to more than 15 days in the prior 3 months. She was drinking soda more frequently because she was tired from staying up so late trying to catch up with missed schoolwork, and she had stopped exercising.
This patient’s CDH was consistent with NDPH. Her treatment required a multimodal approach, which began by reassuring the patient and her parents that there was no evidence of a tumor or any life-threatening neurologic disease and that the arachnoid cyst was incidental, common, and unrelated. They were given information about NDPH and instructions for routine daily activity, sleep, and dietary changes, including avoidance of caffeine, chocolate, nitrates, and monosodium glutamate. Because of the tenderness in the patient’s neck musculature, she was prescribed physical therapy. An exercise routine was started, and pharmacotherapy was initiated with a trial of amitriptyline. Counseling was begun, and behavioral techniques focused on treating stress were discussed. The patient’s parents were instructed not to allow her to miss school because of headache. She was told to limit the use of over-the-counter medications to 2 days/week or 10 days/month. Alternative therapies including supplements, acupuncture, and yoga were discussed. Follow-up was arranged in 6 weeks.
Clinically, NDPH is difficult to treat, and patients with this disorder must be followed up closely. Nonpharmacologic treatments are integral to optimize the possibility of a good outcome. These treatment include confident reassurance, patient and parent education, institution of a regular schedule for eating and sleeping, a good diet with plenty of fluids, counseling for the patient and his/her family, avoidance of school absences, identification of stressors, discontinuation of over-the-counter medications, and initiation of other behavioral strategies, such as biofeedback. If the patient is overweight (ie, has a body mass index > 25), a weight loss program should be initiated. Parental enmeshment and psychopathology may contribute to CDH; therefore, parental counseling may be required.
Pharmacologic therapies for NDPH are the same as those used for other CDH types and for migraine prevention. The most commonly used medications include the tricyclic antidepressants (amitriptyline) and antiepileptics (topiramate, valproic acid, gabapentin). For amitriptyline, we use a starting dose of 5 mg and 1 mg/kg per day as an initial target dose. For teens and preteens, we typically begin with 5 mg at bedtime and increase the dosage by 5 mg every 2 weeks to a goal of 20 mg at bedtime. The patient is then seen in follow-up, and the dosage is adjusted. In thin patients, we often use cyproheptadine because it also increases appetite and improves sleep. We begin with 2 mg at bedtime and increase the dosage by 2 mg every 2 weeks to a goal of 6 mg at bedtime. We frequently use melatonin, 3 mg taken 2 hours before bedtime, in patients who have difficulty falling asleep and titrate the dosage to 9 mg as needed. Additional treatments with reported benefit, but insufficient data, include propranolol, selective serotonin reuptake inhibitors, muscle relaxants, and indomethacin. Botulinum toxin is reported to be safe and effective compared with both placebo and topiramate for CM in adults [30•, 31], and successful treatment of intractable migraine and CDH in adolescents also is reported . These findings suggest a potential role for botulinum toxin in treating NDPH, and success has been reported for this indication in adults [33•].
Alternative therapies reported to be useful include riboflavin, butterbur, coenzyme Q10, magnesium, massage, acupuncture, exercise, physical therapy, weight loss, and yoga. At our center, patients who are refractory to the aforementioned treatments are offered a 3-week intensive multimodal inpatient rehabilitation program.
An interesting but unconfirmed study reported elevated CSF tumor necrosis factor-α (TNF-α) levels in NDPH and treatment-refractory CM . Doxycycline, a TNF-α inhibitor, was given to four patients with refractory NDPH and elevated CSF TNF-α levels who had failed at least five preventatives, infusion therapy, and inpatient treatments . Treatment consisted of doxycycline, 100 mg by mouth twice daily, in an open-label fashion. All patients had a positive response to treatment: two became pain-free, one had 80% improvement, and one had a greater than 50% reduction in frequency of severe pain episodes. Onset of action was about 2 months, although one patient responded within 2 weeks.
Data regarding prognosis in NDPH are sparse, and more data are required. It has been suggested that there are two forms of NDPH. One typically resolves without therapy within several months, and the other is a more refractory form resistant to aggressive treatment [3, 23]. Unfortunately, the refractory form seems to be the most frequently encountered in our practice. In his original paper, Vanast  appeared to describe the milder, “benign” form, in which 73% to 86% of patients have headache resolution in 3–24 months, regardless of treatment . From our experience, risk factors for chronification of NDPH include female sex, straight-A report cards, excess extracurricular activities, poor sleep, a disordered home life, medication overuse, obesity, caffeine, poor diet, stressful life events, head injury, and insufficient exercise and fluids. In general, 30% of children and adolescents seem to improve in 3 months to 4 months, even without intensive treatment, and the rest continue to have CDH indefinitely. Still, 78% may be significantly improved within 2 years .
NDPH is a form of CDH that may combine features of both migraine and tension-type headache. In contrast to other CDH types, NDPH often has the characteristic clinical feature of the patient being able to recall the specific date his or her unremitting headache began. Given this characteristic history, along with overlapping migrainous and tension-type features, consideration should be given to some revisions of NDPH criteria. Many NDPH patients have more than one migrainous feature, although the criteria allow for only one. This makes diagnosis by strict criteria more difficult and leads to diagnostic uncertainty for both physicians and patients. In addition, after several months, NDPH, CTTH, and CM have similar phenotypes and are hard to differentiate. NDPH has an abrupt onset and is unremitting from the start, whereas CTTH and CM typically have a steadily increasing frequency before occurring daily. For this reason, diagnosis of NDPH can be made earlier than the criteria’s minimum of 3 months’ duration of headache. A minimum of 1 month would be more reasonable for NDPH, because its onset is so characteristic. More patients would then fulfill the diagnostic criteria earlier, allowing them to receive adequate treatment sooner. The current 3-month criterion leaves patients with suspected NDPH in a quandary with an uncertain diagnosis and anxiety while they wait to meet the criterion, when they could be starting the multimodal approach required to optimally treat NDPH and to prevent further comorbidities.
Patients with NDPH often undergo extensive evaluations, most of which are negative and unremarkable. The disorder often is difficult to treat and requires a multimodal approach. There are still many unanswered questions regarding NDPH; further study is necessary to better understand, characterize, diagnose, and treat it. Prospective, multi-institutional studies are warranted to better define the clinical features and etiologic factors (physical, psychological, environmental, or others) of this disorder and to determine the value of various diagnostic tests. Studies also are needed to compare nonpharmacologic, alternative, and pharmacologic treatment strategies and to better identify prognostic factors while eliminating institutional or regional bias.
Dr. Rothner received a grant from GlaxoSmithKline for research on Treximet in adolescents. No other potential conflicts of interest relevant to this article were reported.