We conducted a prospective study of a consecutive sample of 244 patients observed in our tertiary level headache centre after direct referral from our hospital’s ED where they presented with a chief compliant of headache. Our aim was to analyzed the different steps of the ED management of patients with headache to detect those deficiencies that can be overcome by a prompt referral to a headache clinic.
The first step of this journey is represented by the need to make the correct diagnosis. The evaluation of patients with headache in the ED environment presents significant differences compared to the evaluation of patients with headache in the clinic. The management of headache in an acute environment should focus on the exclusion of serious secondary causes, on the correct diagnosis, on the management of symptoms and possibly on establishing a continuing care plan [24]. The sense of urgency to rule out secondary and often life-threatening causes of headache, also due to of time constraints due to the need for efficient patient management, can lead ED physicians to interrupt the diagnosis process on the differential diagnosis between primary and secondary headache. Furthermore, a Canadian study to evaluate the diagnostic accuracy of neurological complaints in ED found that in 35.7% of cases the initial ED diagnosis was not in accordance with the final diagnosis possibly made by a neurologist; among the most common problems diagnosed incorrectly was primary headache [25]. There was a high discrepancy rate between the diagnosis made in our headache centre and that made in ED. This discrepancy can have several reasons. ED physicians may be more interested in ruling out serious causes of headache than in distinguishing different primary headaches or they may simply not be familiar with the ICHD diagnostic criteria. However, despite being aware of the diagnostic criteria, ED physicians may not have mattered because they intended to use the same treatment regardless of the diagnosis of primary headache. In addition, there is a possibility that ED physicians may simply accept the diagnosis of a triage nurse or that they may not change the diagnosis already entered in the system. However, the interesting finding was that ED physicians used only four diagnostic categories while triage nurses used eleven different diagnosis and the analyses of concordance among the three different diagnoses (triage, ED physician and headache centre) shown a significant moderate agreement only for the diagnosis of migraine and only between triage and headache centre.
The second step of the journey is the diagnostic investigations that may be necessary to confirm the diagnostic intuition and establish consistent treatment. Head CT is commonly used to evaluate headache due to its prompt accessibility and diagnostic accuracy, but unnecessary head CT scans lead to a longer duration of ED stay [26], exposure to radiations [27] and increased medical costs [28]. The use of CT for non-traumatic headache in the US emergency departments has doubled in the past 20 years. A recent study showed that up to 31% of patients who had headache underwent imaging [11, 29] and head CTs accounted for almost the 50% all CTs performed in United States ED visits [30]. However, of the CT scans performed in the United States EDs among patients with headache between 1992 and 2001, almost 95% showed no pathology [11]. We found that 75% of the patients who attended our ED with a major compliant of headache underwent CT scan which in 93.4% of the cases was negative while in the other cases it showed results all considered accidental and not potentially lethal (Table 3). However, this large use of CT in the ED requires serious reflection. In fact, a recent study found that a 9.6% decrease in the use of head CT for patients who come to ED with a chief complaint of headache was not followed by an increase in death or missed diagnoses [31], adding to the convincing evidence that there is a possibility to safely decrease CT scan for ED patients. In contrast, in support of CT use, another study found that patients who underwent a CT scan of the head during an initial ED visit were about half as likely to return to ED within 30 days than those weren’t underwent CT [32]. Cranial imaging has been shown to have a positive influence on patients’ fear and anxiety levels [33] and, reassuring patients, CT scan prevent return ED visits. However, we can speculate that a rapid referral to a headache unit may have, at least in some patients, a similar reassuring effect but avoiding the impact of CT on the patient and the healthcare system.
The third step is the request for specialist consultations aimed at adding experience to the work of ED physicians. In our ED at least one neurologist is available 24hours a day, 7 days a week, both for first-line evaluation if the pre-hospital evaluation suggests a neurological condition, and for the second-line, according to the judgement of a non-neurologist ED physician at the arrival of the patient. However, less than a fifth (19.3%) of the patients received a neurological consultation (Table 3). This rate was lower than that found in another study conducted in French EDs in which a third of patients received a consultation with a neurologist [8]. The authors speculated that this may have been due to the fact that patients were experiencing an unusually severe attack or because a differential diagnosis was deemed necessary for patients who had reported a first episode of headache. We can assume that the lower rate of consultations in our study may have several reasons and, among these, the doctor’s certainty of excluding a secondary headache and the knowledge of the possibility of a rapid referral to our headache clinic. Further consultations were requested for 8.2% of the patients and the three most frequently consulted specialists were ophthalmologists (for the aura), otolaryngologists (for sinusitis) and psychiatrists (for comorbid conditions). However, having received specialist consultations, both neurological and other, was the only factor, together with the complaint of the first aura in life, to be significantly associated with a longer stay in ED (Table 4). It is possible that in front of a patient who reported having previously had other episodes of aura, the ED doctor was reassured on the benignity of the symptom while in the case of the first aura the doctor had a more cautious attitude and preferred to seek advice. this fact could also indicate an inadequate knowledge of the diagnostic criteria of primary headaches in general, and specifically of the clinical spectrum of migraine.
The fourth and fundamental step in the management of headache in ED is pharmacological treatment. Patients presenting with migraine or other headaches have the third highest self-reported pain scores among all patients presenting to ED with a painful condition [34]. Consistently, the American Migraine Prevalence and Prevention (AMPP) study found that unbearable pain was the patients’ most common reason for using ED for migraine [35]. Patients attending ED with a headache pattern similar to previous migraine attacks generally do not require, or either want, diagnostic tests, but expect rapid and effective management of their headache and any incorrect diagnosis can lead to a generalized pain treatment rather than of migraine-specific treatment [17]. More than 20 different drugs and drug combinations are used to treat migraine in ED, including migraine-specific drugs (e.g., sumatriptan and dihydroergotamine), dopamine antagonists (e.g., metoclopramide, chlorpromazine and prochlorperazine), NSAIDs, opioids, corticosteroids, and anti-histamines (e.g., diphenhydramine and promethazine) [36]. In our ED, the most commonly prescribed pharmacological agents were NSAIDs and weak opioids, administered in 44.3% and 17.6%. of patients, respectively, followed by paracetamol which was administered to 8.2% of patients (Table 3). Treatment for associated symptoms included antiemetics and anxiolytics, administered to 12.7% and 4.9% of patients, respectively. Another study conducted in Europe showed an opposite use of NSAIDs (42.9%) and non-opioid analgesics (61.2%) and a lower use of antiemetics (8%) and anxiolytics (3%) [8]. In the study mentioned above, approximately 9% of patients did not receive any pharmacological treatment and triptans were not administered as frequently (11.2%) [8]. In our sample, approximately 43% of patients did not receive any drugs and no patients received triptans (or ergotamines). However, the limited use of triptans in ED is not limited to European countries. A survey conducted in a United States ED found an equally high percentage of patients (38%) who received neither drugs nor intravenous fluids [17]. Significant variations exist in managing headache between and within EDs due to lack of strong recommendations, physician comfort and familiarity with specific medications, beliefs about efficacy, concern about short-term side effects and patient request [36–38]. Only recently, the Canadian Headache Society and the American Headache Society provided evidence-based therapeutic recommendations for migraine that require treatment in emergency settings [39, 40].
The fifth and final step in the management headache in ED is represented by discharge. Significant evidence suggests that migraine discharge management is often suboptimal. About 40% of emergency physicians never prescribed triptans at discharge [41], approximately 60% of patients had no documented discharge medications and 2/3 of patients did not receive a physician follow-up recommendation [17]. These malpractices contribute to the high rate of return visits to ED noted among migraine patients [42]. Another study found a high percentage of patients (34.7%) discharged without a prescription and also in this case the most commonly prescribed drugs were analgesics and NSAIDs rather than migraine-specific drugs [8]. Our data, in line with the previous literature, shown that 43.9% of patients were discharged without a prescription while the most commonly prescribed drugs for headache treatments were analgesics NSAIDs (33.2) and paracetamol (11.1%) while triptans were prescribed only to 1.6% (Table 5).
Considering what has been previously discussed about the journey through ED of a patient complaining of headache, a fast track between ED and a headache clinic represents a step forward in managing headache. Headache is the most common disorder among patients presenting to ED with neurological complaints [43] and although in most cases it is discharged as non-urgent complaint, those patients feel they need urgent medical evaluation [44]. In contrast, migraine patients, especially those with a known diagnosis, should have an idea that their headache is not a life-threatening condition and if they get to ED it is more because of access issues and less because of a perceived need for emergent treatment. Inappropriate ED utilization causes overcrowding and consequently an increase in intervention times, which is one of the main reasons for leaving ED to prematurely [45], especially among patients with headache compliant [46]. This suggests that a healthcare system could reduce ED visits among migraine patients by increasing the availability of urgent care or walk-in appointments [47].
Another proof of the importance of rapid referral to a headache centre comes from an analysis of the use of ED by patients with migraine. The AMPP study showed that only a small subset of respondents used ED to treat severe headaches in the previous 12 months: 3% visited the ED once and another 3% visited ED more than once [35]. Interestingly, in a previous study, we showed that among 548 migraine patients who attended our headache clinic and who were in continuous treatment and underwent regular follow-up visits, only 0.9% of them entered the ED during the entire 2-years observation period [48]. This finding is explained by the fact that migraine prevention and prescription of migraine-specific rescue therapies reduce resource use in general, and in particular ED visits [49–51]. In addition to improving the health and quality of life of patients, a preferential between ED and the headache clinic can lead to significant economic savings considering that ED care is considerably more expensive than both hospital outpatient care and office-based outpatient care. In a time characterized by economic difficulties and the availability of new and expenses treatments, such as the monoclonal antibodies targeting the calcitonin gene-related peptide, a virtuous allocation of economic resources becomes mandatory [52].