The applicability research of the diagnostic criteria for 6.7.2 angiography headache in the international classification of headache disorders-3rd edition

Previous studies have reported that the incidence of headache in cerebral angiography ranges from 6.9–55.6% [1‐6]. The possible mechanisms for angiography related headache are irritation of the trigeminovascular system by contrast agents or mechanical stimuli such as the catheter itself, activation of platelets or irritation of the vascular endothelium, and selective relaxation and dilation of intracranial vessels [1, 2, 7‐10]. ICHD-3 version published in 2018, the diagnostic criteria for AH (Chapter 6.7.2) is showed in Table 1. The criteria defined the onset of cerebral AH to no more than 24 h after cerebral angiography and headache resolved within 72 h after the angiography. In clinical practice we found that headache duration was less than 24 h in many patients with cerebral AH. Nearly half of patients still had headaches after 24 h of cerebral angiography, and about a third of patients have more than one headache after cerebral angiography. These phenomena did not meet the diagnosis criterion of cerebral AH in ICHD-3. In addition, Ramadan proposed the mechanism of headache may be that the catheter or the contrast material activate platelets or damage the vascular endothelium thereby resulting in the release of vasoactive substances such as nitric oxide (NO) or serotonin (5-HT). Should this hypothesis be true, the peripheral or coronary angiography will have the same headache incidence as the cerebral angiography [2].

Based on the above points, in order to further discuss the characteristics and potential mechanisms of cerebral AH and improve the applicability of cerebral AH diagnostic criteria in ICHD-3, we designed this prospective study.

This project was approved by the Research Ethics Committee of the Chinese People’s Liberation Army (PLA) General Hospital. All of the patients provided written informed consent.

The study included three groups from the First Medical Center of PLA General Hospital, respectively patients admitted to the Department of Neurology and the Department of Cardiology from May 2020 to January 2021, patients admitted to the Department of Vascular and Endovascular Surgery from October 2020 to January 2021. The inclusion criteria were: 1. Older than 18 years of age; and 2. Cerebral angiography or coronary intervention or extremities arterial intervention under local anesthesia. The exclusion criteria were as follows: 1. Patients with subarachnoid hemorrhage, patients who presented with secondary headaches due to the reason for DSA or had a headache occurring within 24 h prior to the procedure; 2. Renal failure requiring dialysis; 3. Severe organ failure, unconsciousness or endotracheal intubation; and 4. Patients who underwent intervention under sedative or general anesthesia; 5. Refusal to complete the questionnaire or follow up regarding the questionnaire.

All patients underwent preoperative serum biochemical tests, blood coagulation, electrocardiogram, chest radiographs or chest CT, and fasting for 8 h and drinking abstaining for 4 h before the procedure. All procedures were performed by senior interventional specialist with the assistance of residents. Cerebral angiography used one or two of Iodoxanol, Ultravist, loversol, lohexol and lopamidol as contrast agents. Coronary intervention includes coronary angiography (CAG) and percutaneous coronary intervention (PCI). Iodixanol or Ultravist was used as contrast agent, and a small amount of nitroglycerin was used during coronary intervention. Extremities arterial intervention consisted of extremities arteriography, balloon dilation, stent implantation, and peripheral aneurysm repair, using iodixanol as contrast agent.

A detailed headache questionnaire was designed by our team based on the ICHD-3 diagnostic criteria. The questionnaire consists of three parts: 1. Basic demographic information and medical history of the patient. 2. Information of interventional procedure. 3. Headache information: (a) Preoperative visit (information of previous headache history): headache age, attack frequency, duration, headache location, quality, intensity, accompanying symptoms, alleviating factors, family headache history, treatment, etc. (b) Postoperative visit (postoperative follow-up time points): headache occurrence, onset time and duration of headache, headache location, quality, intensity, accompanying symptoms, treatment, etc. Intensity was assessed with a numeric rating scale, which is a segmented numeric version of the visual analogue scale (VAS), in which the patients select a whole number from 0 to 10 that best reflects their pain level [11].

The diagnosis of headache is mainly based on the clinical characteristics of headache, such as headache onset pattern, frequency, duration, location, quality, intensity, accompanying symptoms and alleviating factors. In ICHD-3, the diagnostic criteria for 6.7.2 angiography headache do not describe the location, quality, intensity and accompanying symptoms of headache (Table 1). In ICHD-3 beta version, the diagnostic criteria for angiography headache (Chapter 6.7.2) have only been based on three studies [12].The diagnostic criteria for angiography headache in ICHD-3 were based on five studies, only one study included peripheral angiography as a control group [13].

Of the 5 studies, all patients included in Shuaib et al. ‘s study had a history of migraine, and their study focused on the relationship between migraine and neurological complications after angiography [14]. The study of Ramadan et al. only included 45 patients, with a small sample size and a short follow-up time [2]. The study of Gil-Gouveia included 32 patients with subarachnoid hemorrhage, and only 48 patients after subarachnoid hemorrhage was removed [1]. Gündüz’s study mainly focused on headaches related to angiographic therapeutic procedures [3]. The study of Aktan was a prospective case-control study, including 139 cases of cerebral angiography and 30 cases of peripheral angiography. However, the sample size of the control group in the study was only 30 cases, which was significantly less than that of the experimental group. Therefore, the incidence of headache in the two groups could not be statistically analyzed, and the characteristics of headache in the control group were not described. In addition, in Aktan ‘s study, patients were informed before angiography of the possibility of headache during and after angiography, which may bring mental stress and thus affect the incidence of headache [4].

Our study showed that the incidence of headache in cerebral angiography was 22.4% (24/107), 23.8% (24/101) in coronary intervention and 16.9% (12/71) in extremities arterial intervention. The incidence of headache in the cerebral angiography and coronary intervention group was similar, slightly higher than that in extremities arterial intervention group, and there was no statistical significance in the incidence of headache in the three groups.

In our study, the incidence of headache in coronary intervention group (23.8%, 24/101) and extremities arterial intervention group (16.9%, 12/71) was higher than the incidence of headache in peripheral angiography group of Aktan (10%, 3/30) [4]. This may be due to that the control group only had 30 patients, which could easily cause bias. There were a few studies on cerebral angiography headache, and the incidence rate was different in the previous reports, which may be caused by different studies had different study methods. In terms of research methods, for a prospective cohort study, we set three groups: cerebral angiography group, coronary intervention group and extremities arterial intervention group. In order to reduce patient recall bias, we set five follow-up time points: immediately after the procedure, 24 h, 72 h, 1 week and 2 weeks after the procedure. During the exclusion criteria and follow-up, we excluded factors which might influence the incidence of headache, for example, not informing patients of the possibility of headache related to the intervention before the procedure, avoiding psychological suggestion to patients, and thus reducing headache caused by mental factors such as anxiety [15, 16]. We excluded patients had headache within 24 h prior to the procedure in order to avoid interference with the procedure related headache. Patients with renal failure requiring dialysis may have 10.2 dialysis headache and were excluded [13, 17]. The above experimental scheme design improved the quality control of our study.

In our study, female was not a risk factor for headache in coronary intervention group and extremities arterial intervention group, but a risk factor for cerebral angiography headache, which is consistent with the research of Gil-Gouveia [1]. The possible explanation is that women demonstrating lower pain thresholds and the fluctuations of sex hormones especially estrogen and progesterone may cause a change in the prevalence or intensity of headache [1, 18, 19]. Previous history of headache as a risk factor for angiography headache had been established in many studies [2‐6, 20]. The possible mechanism maybe that some individuals with a history of headache had lower thresholds for trigeminal activation [2, 7]. Although previous literature had reported that patients with cerebral aneurysms or cerebral infarction were prone to headache [13, 21], our study and previous studies showed that the occurrence of cerebral angiography headache was not related to these angiographic indications [2, 3]. High education level was a risk factor for angiography headache in study of Kwon. They suggested that it was possible that patients with high education level were more anxious and/or nervous with the procedure [5]. In our study, beyond high school was not a risk factor for cerebral angiography headache, possibly because education level could not represent mental state. Future researches should study education level together with related mental scales such as anxiety and depression. We studied the incidence of angiography headache of the two procedure approaches namely transradial and transfemoral, the results showed that there was no statistical significance between the two procedure approaches. Other factors such as age, BMI, allergic history, tobacco and alcohol history, hypertension, diabetes, cerebral infarction, hyperlipidemia, coronary heart disease, intervention history, and whether complicated with severe vascular stenosis were not risk factors for cerebral angiography headache.

Previous study had shown that the occurrence of migraine after cerebral angiography was related to meningeal vasodilation [9]. Studies had reported that patients underwent cerebrovascular interventional therapy had a higher incidence of headache than cerebral angiography [3, 7]. The occurrence of headache during cranial embolization has been attributed to vessel distension occurring with balloon inflation, which stimulates pain receptors in the vessel wall, producing focal areas of referred pain [22‐24]. Theoretically, it can also occur with contrast injection, if the pressure within the vessel is enough to reach the threshold of mechanical stimulation of pain receptors [7]. In the cerebral angiography group and coronary intervention group of our study, the intraoperative headache occurred within 1 h after the beginning of the procedure, the duration time was less than 1 h. Most of the patients had mild to moderate throbbing headache, and the headache characteristics were similar to 8.1.1.1 immediate NO donor-induced headache in ICHD-3 [13, 25, 26]. There were no statistical differences in onset time, duration and VAS of postoperative headache among cerebral angiography group, coronary intervention group and extremities arterial intervention group, and most of the headaches were throbbing, suggesting that the cause of headache might be as hypothesized by Ramadan-the catheter or the contrast material activate platelets or damage the vascular endothelium thereby resulting in the release of vasoactive substances such as nitric oxide (NO) or serotonin (5-HT) which have been implicated in the pathogenesis of vascular headache [2]. In our study extremities arterial intervention group had no intraoperative headache. The probable explanation is that during cerebral angiography and coronary intervention, catheterization and/or contrast agents pass through the brachioid trunk and the aortic arch, damaging and inducing vascular endothelial cells to produce vasoactive substances (such as NO, 5-HT, etc.). These vasoactive substances may enter the brain via the carotid and/or vertebral arteries and cause headache. In our study, 97.2% (69/71) of patients in extremities arterial intervention group underwent lower extremity arterial intervention and 91.5% (65/71) were transfemoral approaches. Catheters and contrast agents mostly did not pass through the brachioid trunk and the aortic arch. In addition, since the lower extremity arteries are relatively far away from the brain, the vasoactive substances produced by the injury-induced vascular endothelial cells during the interventional procedure are rapidly metabolized and absorbed instead of entering the brain. The mechanism of cerebral angiography headache is unknown, and no related animal experiments have been conducted to explore the mechanism. In the future, more studies are needed to pay attention to the mechanism.

In ICHD-3, diagnostic criterion A in 6.7.2 is “Any new headache fulfilling criterion C”. criterion C1 is “headache has developed during or within 24 hours of the angiography”. Both our study and previous reports have confirmed the headache during the procedure [3, 4, 7]. In the study of Aktan [4], of all patients who developed angiography headache, 23.8% (10/42) was during the procedure, 61.9% (26/42) was within 24 h after the procedure and 45.2% (19/42) was within 2–7 days after the procedure. The onset time of headache in these 45.2% patients was not within the time range defined by criterion C1. In our study, headache still occurred in 13 patients within 2–14 days after the procedure, accounting for 54.2% (13/24) of the total headache patients, and these 54.2% patients was also not within the time range defined by criterion C1. Figure 4 shows the cumulative frequency distribution of patients with cerebral AH at each follow-up time point in our study. Patients with cerebral AH during the procedure was 33.3% (8/24), up to 24 h after the procedure, the patients with cerebral AH accounted for 62.5% (15/24) of the total number of patients with cerebral AH, up to 3 days after the procedure and up to 7 days after the procedure, the patients with cerebral AH accounted for 70.8% (17/24) and 87.5% (21/24) respectively. According to our results, C1 criteria only included 62.5% of patients with cerebral AH. If the time specified in the criterion C1 is extended to 1 week after the procedure, 87.5% of patients with headache can be included, which will significantly improve the sensitivity of the criterion C1. Therefore, we suggest that the criterion C1 should be changed to: headache has developed during or within 7 days of the angiography.

Criterion C2 defines as “headache has resolved within 72 hours after the angiography”. In our cerebral angiography group, the average duration of headache during the procedure was (8.27 ± 11.20) min, with a median of 3 min; The average duration of headache within 24 h after the procedure was (400.00 ± 363.90) min, with a median of 420 min. The average duration of headache within 2–14 days after the procedure was (123.72 ± 200.77) min, with a median of 30 min. Aktan reported that the mean duration of headache during angiography was (41.7 ± 65.4) min, with a median of 22.5 min. The mean duration of headache within 24 h after the procedure was (171.7 ± 208.1) min with a median of 75 min, and the mean duration within 2–7 days after the procedure was (122.1 ± 173.1) min with a median time of 60 min [4]. Kwon reported that 77.9% of headaches resolved within 24 h [5]. The criterion C2 is defined as headache resolved within 72 h after the angiography. The time span is obviously too long, so we recommend that criterion C2 be modified to: headache resolved within 24 h of a single headache attack.

Criterion C3a defines as “developing during contrast injection and lasting <1 hour”. In our study, a total of 8 patients developed headache during angiography, with an average duration of (8.27 ± 11.20) min and a median duration of 3 min. Aktan reported [4] that the mean duration of headache during angiography was (41.7 ± 65.4) min, with a median of 22.5 min. Gündüz reported [3] the duration of headache during angiography was less than 10 min. Gil-Gouveia Rreported [7] the duration of headache during angiography varying from 2 to 30 s. These findings support criterion C3a, therefore, it is recommended that criterion C3a should be retained.

Criterion C3b defines as “developing a few hours after the angiography and lasting >24 hours”. The criterion C3b and the criterion C3a are in parallel relationship. The criterion C3a is for the characteristics of headache during angiography, while the criterion C3b is obviously for the characteristics of headache after angiography. As in the previous discussion of criterion C1 and C2, the description of criterion C3b that headache developing a few hours after the angiography and lasting >24 h is obviously not appropriate. It is suggested to be modified to “ developing a few hours to 7 days after angiography and a single headache lasting < 24 hours”.

Criterion C3c defines as “occurring in a patient with 1. Migraine and having the features of 1.1 Migraine without aura or 1.2 Migraine with aura” In our study, some patients with a history of migraine developed angiography headache with the same characteristics as their previous migraine, and some patients had visual aura, which is supported by previous reports [2, 6]. It is recommended that criterion C3c should be retained.

Criterion B defines as “Intra-arterial carotid or vertebral angiography has been performed”. In our study, we found that:

Based on the above analysis, we believe that cerebral angiography headache, coronary intervention-associated headache and extremities arterial intervention -associated headache have similar headache characteristics, suggesting that the occurrence of this kind of headache is not limited to whether intra-arterial carotid or vertebral angiography is performed, coronary intervention and extremities arterial intervention can also lead to similar headache. Criterion B seems to be expanded from “intra-arterial carotid or vertebral angiography has been performed” to “intra-arterial carotid or vertebral angiography or coronary intervention or extremities arterial intervention has been performed”. Considering that 6.7.2 angiography headache in ICHD-3 belongs to classification 6, namely “Headache attributed to cranial and/or cervical vascular disorder”, coronary intervention and extremities arterial intervention do not belong to the category of Cranial and/or cervical vascular disorder. Therefore, we recommend that ICHD add categories of “headache due to coronary intervention” and “headache due to extremities arterial intervention”.

The note of 6.7.2 angiography headache in ICHD-3 states that “there are three recognized (but not separately coded) subforms of 6.7.2 Angiography headache”.

This note is discussed in criterion C3c and we recommend keeping this note.

Previous studies showed [2, 4] cerebral angiography headache is mostly mild to moderate, throbbing or pressing pain. In our study, cerebral angiography headache was mostly mild-moderate, throbbing or pressing pain, and some patients were accompanied by photophobia, photophobia and visual aura, accompanied by nausea and no vomiting. It is suggested to add a description of the characteristics of headache to the diagnostic criteria, such as: mild-moderate, throbbing or pressing pain with photophobia, nausea and no vomiting. This will increase the clinician’s ability to identify this kind of headache.

Aktan reported 31.0% (13/42) of patients had more than one headache attack after cerebral angiography, and Gil-Gouveia reported that 48.8% (40/82) of patients had recurrent headache [1, 4], The same phenomenon was found in our study, with 41.7% (10/24) patients of cerebral angiography group having two or more headaches. So, we suggested to add to the diagnostic criteria: about one-third of patients may have more than one headache attack.

Based on the above discussion, we suggest that the new diagnostic criteria for 6.7.2 angiography headache in Table 14.