Abstract
Background
There is no defined preventive treatment protocol for persistent post-craniotomy headache. In several small case series and individual case reports onabotulinumtoxinA injected into the craniotomy scar has shown possible efficacy. What is lacking is long term follow-up and if focusing on the cranial suture lines along with the craniotomy scar can enhance improvement and provide more sustained benefit.
Methods
Retrospective chart review with case series.
Results
Four patients (three women, one man) with ICHD-3 defined persistent post craniotomy headache were treated using a novel onabotulinumtoxinA injection protocol. All the patients presented with continuous head pain of moderate to severe intensity. All had severe allodynia on the side of their craniotomy. All had significant reduction in quality of life. Our application of onabotulinumtoxinA involved injection into both the surgical scar and the transected/irritated cranial suture lines noted on neuroimaging and physical examination. With treatment all patients demonstrated significant benefit including a reduction in daily pain intensity (75%-100%), developing periods of pain freedom (2–7 days per week) and having a dramatic improvement in quality of life (close to 100% in all). The benefit was sustained for at least five years of follow-up.
Conclusion
From our case series it appears that injection not only along the painful craniotomy scar but into the involved cranial suture lines provides positive efficacy and sustained improvement in patients with persistent post craniotomy headache.
Introduction
The current International Classification of Headache Disorders (ICHD-3) criteria defines both acute and persistent forms of headache attributed to craniotomy (1). The head pain can begin within seven days of surgery and either last less than or more than three months in duration. The true incidence and prevalence of persistent post craniotomy headache (PCH) is not well-established but what is known is that chronic head pain developing after a craniotomy procedure can be of moderate to severe in intensity and substantially alter an individual's quality of life (2,3). At present there is no specific preventive treatment paradigm for persistent PCH (3,4). Various medications from non-steroidal anti-inflammatories (NSAIDS), anti-epileptics and opiates have been tried as well as local anesthetic injections, but none show a consistent and effective response (3). There has been recent recognition that onabotulinumtoxinA injections may provide substantial benefit for persistent PCH, but this has come only from individual case reports and small case series (5,–9). However, in many instances the documented patients did not meet ICHD-3 criteria for PCH, the exact location of injections was not established and there was no long-term follow-up. We now present treatment outcome data for a case series of four patients with ICHD-3 defined persistent PCH with many years of continuous therapy. Our treatment protocol was unique in that we firstly defined PCH not only by the location of the craniotomy surgical site and residual scar but also by what cranial suture lines were disrupted directly and indirectly during surgery. Then we utilized a combination injection technique both to the craniotomy scar and to the transected/irritated cranial suture lines.
Methods
A retrospective chart review was carried out identifying patients with persistent PCH who had been treated with our combination onabotulinumtoxinA injection protocol consistently for one or more years in duration. Four patients were identified. Patient demographics are in Table 1. There were two other cases documented on chart review, but they only received cranial suture injections because of post-surgical skull anatomy issues. These outlier cases will be presented in the Results section but are separate from our studied combination injection protocol patients. Each was initially seen in our academic headache clinic during the years 2015–2020. A diagnosis of ICHD-3 defined persistent PCH was made at the initial consultation. The patients were deemed medicinal treatment refractory failing at least two headache preventives prior to starting onabotulinumtoxinA treatment. All presented with persistent daily pain without any pain free time. Initial examination focused on head pain location as defined by the patient with direct palpation over the craniotomy site, craniotomy scar and cranial suture lines within the surgical field. In addition, we also looked for suture irritation beyond the surgical area, so we examined the main cranial sutures bilaterally including sagittal, coronal, squamosal, sphenoparietal, sphenofrontal and lambdoidal. Suture irritation was defined by the patient experiencing pain when moderately pressing directly over the cranial suture lines. In all instances the patients localized their head pain to their craniotomy site and residual scar and did not recognize the cranial suture irritation until direct palpation. Neuroradiology helped to define what cranial suture lines were transected during surgery (Figures 1–4). In addition to defining palpable pain, areas of allodynia were also noted. Allodynia was defined by the patient noting extreme discomfort (normally with a facial grimace) with light touch of the scalp by the examiner's fingers. Prior to treatment skull imaging was completed looking for any substantial post-surgical skull gaps (which would preclude injections) and the patient's neurosurgeon needed to clear them for onabotulinumtoxinA injection within and/or near the craniotomy site.
A: Frontal radiograph of the skull demonstrates a left frontal craniotomy (arrow), which extends in-plane through the coronal suture. B: Graphic image of skull cranial suture lines lateral view-red line indicates the involved suture lines with craniotomy.
A: Lateral scout image of the skull demonstrates a craniotomy (arrow), which traverses the coronal and squamosal sutures. B: Graphic image of skull cranial suture lines lateral view-red line indicates the involved suture lines with craniotomy.
A: Lateral scout image of the skull demonstrates a craniotomy (arrow) traversing the sphenofrontal, sphenosquamosal, and squamosal sutures. B: Graphic image of skull cranial suture lines lateral view-red line indicates the involved suture lines with craniotomy.
A: Lateral radiograph of the skull demonstrates a craniotomy (arrow) traversing the coronal, sphenofrontal, sphenosquamosal, and squamosal sutures. B: Graphic image of skull cranial suture lines lateral view-red line indicates the involved suture lines with craniotomy. The skull cranial figures are free to use as per author's institution
Patient demographics.
For the procedure itself the goal was to direct onabotulinumtoxinA injections to both the craniotomy scar (injecting into the scar line) and any cranial suture lines deemed irritative and contributing to the PCH based on physical examination and neuroimaging. The amount of onabotulinumtoxinA given was determined by length of the craniotomy scar and the number of cranial suture lines involved and to what extent. We would inject over the entire length of the scar even if some of it was non tender, but we only injected over the part of the suture lines that were painful to palpation. Each patient was asked to provide a percent improvement (0–100%) for both pain intensity and quality of life (QoL). Their assessment of QoL improvement was a subjective perception response only and not defined by a published scale. In addition, they were asked if they had developed any pain free time and if so the number of days per week. We present their current improvement after years of treatment (Table 1).
Ethics
This study was approved by the Mayo Clinic Institutional Review Board-IRB number: 19-002635.
Written informed consent was obtained from the patients for publication of this case series and any accompanying images.
Results
Four patients with persistent PCH and who received the combination injection protocol are presented. Three were women. Craniotomy was for a neoplasm in two patients, an aneurysm and epidural hematoma in the others. Daily persistent headache prior to our onabotulinumtoxinA treatment protocol ranged from five months to three years. All patients received onabotulinumtoxinA treatment for at least three years (range 3–7 years). Follow-up has been for 5–7 years. Of note, all the patients did receive a combination anesthetic (1% lidocaine) and corticosteroid (triamcinolone 40 mg/ml) injection into their craniotomy scar prior to having their first ever onabotulinumtoxinA treatment. In each there was a mild but fleeting improvement (several hours) in head pain but not significant enough to continue this mode of therapy and thus onabotulinumtoxinA was scheduled.
Case 1
A 42-year-old woman with a prior history of episodic migraine since her teens underwent a left frontal craniotomy for removal of a biopsy proven pilocytic astrocytoma involving the left subcortical cingulate gyrus. The craniotomy involved the left frontal and parietal bones (Figure 1A and B). Within 1–2 days post-surgery she recognized a persistent pain along the craniotomy site. Average pain intensity was moderate to severe. When the pain peaked, she would develop migrainous associated symptoms. She was unable to touch the side of her headache as it was painful and sensitive. The present headaches were different than her prior migraines which could occur on either side of her head, with preference to the right side and would only last several hours with abortive medication. She presented for consultation five months after her surgical procedure. On examination she had significant pain to light palpation all along the surgical scar with severe allodynia. In addition, her pain extended along the left coronal suture line and left side of the sagittal suture line. Imaging verified surgical transection of the coronal suture. She was on several psychiatric medications which did not help her headaches and limited the addition of other preventives. Topiramate was ineffective as well as NSAIDs and muscle relaxants. Due to the refractoriness of her headaches onabotulinumtoxinA was recommended. On her initial treatment she received 45 units along the craniotomy scar and over the coronal and sagittal suture lines. By her third treatment she reported a 70%-80% improvement in PCH with significant reduction in daily pain intensity and the development of pain free time. To try to enhance the effect the dose was increased to a total of 140 units over the craniotomy scar and irritated suture lines. She has now undergone 30 total treatments over six years duration at every three-month intervals, although there were several times she had to miss a procedure due to personal issues. At present her post craniotomy pain is greatly improved with an 80% reduction in daily pain intensity and with significantly less allodynia. She has pain free time at least 1–2 days per week. She has had 100% improvement in QoL related to her PCH. Her pain recurs by the time the next treatment is needed.
Case 2
A 51-year-old man with no prior headache history presented two years after having a craniotomy for a left middle cerebral artery (MCA) aneurysm with severe daily persistent headaches in the region of his cranial incision. The headache was present almost immediately after he awoke from surgery. The daily pain was moderate to severe in intensity, and he was unable to touch the left side of his head without extreme discomfort. When the pain became severe, he developed migrainous associated symptoms. He was taking NSAIDs daily with minimal relief. He was limited to preventive medication as he had liver cirrhosis. He did try amitriptyline, topiramate and gabapentin without success. The craniotomy involved the left frontal, parietal and temporal bones (Figure 2A and B). On physical exam there was distinct left sided temporalis muscle hypertrophy and spasm and severe palpable pain and allodynia over the craniotomy scar as well as over the coronal and squamosal suture lines. Neuroradiology verified surgical transection of the squamosal and coronal sutures. He started onabotulinumtoxinA for his persistent PCH. He initially was given 75 units total over the left temporalis muscle, craniotomy scar and involved suture lines. At the time of his second treatment the temporalis muscle spasm had almost fully resolved and there was less allodynia in the region of the craniotomy. After the second treatment he developed pain free time, now having head pain only 1–2 days per week. At this point his left sided allodynia had almost resolved. One could palpate over the craniotomy site without inducing discomfort whereas prior he would actually cry with light touch over the same region. He has now had 30 treatments total over seven years. The average craniotomy/suture directed dose was 120 units. Overall, the treatment has been consistent and effective. It normally lasts about 10–11 weeks although when the pain recurs it never matches his pre- onabotulinumtoxinA baseline. His temporalis muscle hypertrophy has remained resolved and he has very little baseline allodynia. He currently has a 90% reduction in headache intensity with 5–7 days per week of pain freedom. His QoL has improved by almost 100%.
Case 3
A 17-year-old girl with no prior headache history presented with a three-year complaint of daily headache. She was involved in a golf cart accident in which she hit her head, developing a left sided epidural hematoma which required a large left parietotemporal craniotomy (Figure 3A and B). Immediately post craniotomy she developed a daily persistent headache. Her headaches were predominantly left sided following along the distribution of the craniotomy incision. Her baseline pain intensity was moderate, with pain exacerbations periods to severe several times per month which could last days in duration. She would develop migrainous associated symptoms with peak pain. On examination she had significant pain and allodynia along the left craniotomy scar plus pain to palpation over multiple cranial suture lines including squamosal, sphenosquamosal, sphenofrontal, sagittal and coronal. Imaging verified the transection of the sphenofrontal, sphenosquamosal, and squamosal sutures. Prior failed medications included topiramate and gabapentin. She was on multiple psychiatric medications for severe depression thus limiting the addition of other preventives. Her initial onabotulinumtoxinA attempt was impeded by significant scar tissue that had developed in her craniotomy incision. The injection needle would bend and sometimes not even penetrate the scar line. She, however, saw significant improvement after several treatments developing pain free time. She has had a total of 18 onabotulinumtoxinA treatments over three years. Average dosing was 115 units. We lost contact with the patient for about two years but on recent follow-up she was having consistent pain free time with minimal to no post craniotomy pain. Thus, in essence the onabotulinumtoxinA alleviated her PCH syndrome.
Case 4
A 47-year-old woman with no prior headache history presented with a daily persistent headache for the past six months after having a craniotomy for a left sphenoid wing meningioma. The patient underwent two craniotomies over a two-month time-period as the first surgery was not successful at fully removing the neoplasm. It was immediately after the second craniotomy, which involved the left frontal, parietal and temporal bones, that her headaches began (Figure 4A and B). The head pain was entirely left sided involving the craniotomy scar but also radiated into her vertex and supraorbital region. The daily pain was constant, varying between mild to severe in intensity. When the pain exacerbated, she would have some migrainous associated symptoms. She could not put any pressure on that side of the head without significant worsening of her pain. Her headache was so incapacitating she required daily opiates. Medications previously tried with minimal relief included duloxetine, gabapentin, tramadol, cyclobenzaprine and methocarbamol. On examination she had significant pain and allodynia over the craniotomy scar itself but also pain to palpation over the left coronal, squamosal, sphenoparietal, sphenofrontal and sagittal suture lines. Imaging denoted transection of the coronal, sphenofrontal, sphenosquamosal, and squamosal sutures. Because of the intractability of her headaches and multiple failed preventive medications she started on onabotulinumtoxinA treatment. Initially the onabotulinumtoxinA was completed every three months but she would have severe pain exacerbation after eight weeks so the interval between treatments was changed to every two months. Average dosing was 165 units. After three treatment sessions she was completely off all opiates. After 12 sessions we noted that the allodynia over the craniotomy site was significantly reduced. She has now had a total of 28 treatments over a five-year time-period. Her daily pain intensity has been reduced 75% and she has six days per week of pain freedom. Her QoL has improved 100%.
Suture line injection-only cases
We have treated two other patients with persistent PCH, but they did not receive the combination injection protocol. One case is a 60+ year old woman with no prior headache history who presented with a daily headache that began 21 years prior, immediately after a left retromastoid craniotomy for an acoustic neuroma. On exam she had no pain over the craniotomy site but had significant contralateral cranial suture-based pain involving the right squamosal, sphenoparietal, coronal and sagittal suture lines. The assumption was that she developed cranial suture irritation from skull manipulation during surgery and possibly from the head fixation device used during the procedure. Thus, she met ICHD-3 criteria for PCH but when treating her we only injected onabotulinumtoxinA over the irritated suture lines and not to the area of the craniotomy. Of note even if the craniotomy area was painful, we could not inject as there was a skull gap in that region. The patient has been followed for four years receiving treatment every three months. Her average dosage is 125 units. She has had an 80% reduction in pain intensity and can have up to 5–7 days per week of complete pain freedom. Her QoL is 100% improved. Prior to her consultation she had failed numerous preventive medications and even PREEMPT onabotulinumtoxinA without any benefit. The injections, however, were never over her cranial sutures.
A second case is a 40+ year old man with no prior headache history who presented with a daily headache for five years which developed immediately after a right sided temporal craniotomy surgery for epilepsy. He had tried multiple preventives without any benefit. His pain was over his right temple in the area of his craniotomy and involved the sagittal, sphenoparietal and coronal suture lines that were transected with the surgery. The issue was that there was a very large post-surgical skull gap in the region of his maximum pain, thus we could not inject in that surgical zone. He had four cycles of onabotulinumtoxinA (dose was 75–125 units) to the suture lines but found no relief. Of note he also did not respond to anesthetic/corticosteroid injections around the craniotomy region.
Discussion
We present the long-term results of utilizing onabotulinumtoxin A injections for persistent post craniotomy headache in a small case series of patients. Our series is unique as it appears to be the first to denote PCH location not only to the position of the craniotomy site and scar but also to the cranial suture lines that were directly transected by the surgical incision and additionally to the suture lines irritated by the procedure but without direct surgical trauma. Our application of onabotulinumtoxinA involved injection into the surgical scar itself but also into the irritated cranial suture lines. In addition, we present one of the longest observations of onabotulinumtoxinA treatment effect for persistent PCH. All of our patients who had the combination injection protocol did very well. This provided pain free time (three of four patients developed 5–7 days per week of pain freedom while prior they were all medicinal treatment refractory and had constant head pain), but also a significant decrease in pain intensity (75%-100%), alleviation of extreme allodynia and a significant improvement in QoL (close to 100% in all) (Table 1). One of our patients was “cured” by the procedure while the remainder have continued treatment as their pain would recur but not to pretreatment intensity levels. Relief was consistent with long-term follow-up (5-7 years). Our two outlier cases that did not receive the combination injection protocol had mixed responses. One had a very positive effect and that may relate to the fact her maximum pain was over the cranial sutures which we were able to inject with onabotulinumtoxinA, while in the second patient his area of maximum pain was in a region we could not inject because of a skull gap.
At present there are only five other studies that have looked at onabotulinumtoxinA for persistent PCH. All focused on injecting the craniotomy scar. The first was published in 2015 and reported on three patients (5). The exact location of the craniotomies and if cranial suture lines were affected was not mentioned. Dosing varied between 25 and 50 units. All the patients reported improvement in headache. Two patients were followed for one year or less while one was followed for 2.5 years. The issue however is that two of the three patients developed their headaches months after their craniotomy thus not meeting ICHD-3 criteria (1). Another case series included four post craniotomy patients with a dramatic improvement in headache and with no side effects (6). This series evaluated patients who developed post craniotomy pain months to years after surgery thus again not meeting ICHD-3 criteria and the headaches were non continuous and of short duration (10–60 min) questioning the PCH diagnosis. As there was no mention of cranial autonomic symptoms this could have been post-surgical paroxysmal hemicrania or cluster headache. A third report looked at two patients who developed pain after craniotomy for a neoplasm (7). Both were followed for more than eight years with significant improvement in headache (70–75% and 80–90%). Dosing utilized was 175–200 units. One patient, however, did not meet ICHD-3 criteria as her headache began one-month post-surgery. From the reported injection site diagram there was definite involvement of cranial suture lines although the authors did not comment on this. Another study looked at four patients with persistent PCH (8). Dosing ranged from 10–80 units along the craniotomy site. All patients were better with serial injections, but a percentage improvement was not documented. The exact duration of follow-up was not well defined. Treatment location was mentioned but the exact location of the injections and if indeed any cranial suture lines were treated was not noted. Finally, two cases of headache after skull decompression for stroke were recently published (9). However, neither patient met ICHD-3 criteria because of delay from surgery to headache onset (six weeks and three months). Both patients were treated with 100 units, with significant decrease in pain intensity (60%-90%). There was no mention if cranial suture lines were injected, and the duration of follow-up was not well defined.
One distinct finding in one of our patients (Patient 2) was painful temporalis hypertrophy/spasm post craniotomy. This was alluded to in a prior case series in which four out of four patients developed temporalis hypertrophy and all improved with onabotulinumtoxinA (6). Injection directly into the spasmed muscle of our patient was essential for pain improvement and after several rounds of treatment this completely alleviated and never returned with years of follow up.
The exact mechanism by which persistent PCH develops is unknown. Several hypotheses include nerve entrapment within the post craniotomy scar itself, dural traction with muscle adhesion to the meninges, and the development of neuromas within the surgical scar (2). Could post craniotomy headache be an entirely extracranially generated head pain disorder? It has been only in the last decade that the pathogenetic construct of extracranial head pain syndromes has come to be identified. Trigeminal sensory nociceptors innervating the scalp tissue and muscles as well as the skull periosteum are potential targets of injury with craniotomy procedures. In addition, researchers have identified a network of sensory and pain fibers within the cranial sutures of the skull which may originate from dural nociceptor axons (10). Irritation of these cranial suture trigeminal nociceptors may be necessary to develop PCH (11). These headaches are possibly less common when the craniotomies involve calvarial bones that are remote from cranial suture lines, but this has not been well defined in prior epidemiologic studies. All of our patients developed migrainous associated symptoms with their PCH although only one patient had a prior migraine history and her headache after surgery was different in presentation than her previous migraines. It has been suggested that skull based trigeminal nociceptors can send signals centrally to the sensory trigeminal nucleus caudalis which could then produce a migraine like headache even though the generator is extracranial (10,12). Of note in the recently defined ‘cranial suture headache’ almost one-third of patients developed migrainous associated symptoms even though their syndrome was purely extracranial based (10,13).
The exact mechanism by which onabotulinumtoxinA works in chronic headache is still not completely understood. The presumed mechanism of action is central modulation at the brainstem and spinal cord level but there is emerging evidence of an additional peripheral effect (12). In many instances our treated patients showed benefit without injection into muscle and many of the typical locations used in the chronic migraine PREEMPT paradigm. Pain relief appeared to be from direct onabotulinumtoxinA administration into the craniotomy scar and transected/irritated cranial suture lines. One of the authors has recently defined cranial suture headache (not secondary to craniotomy) and the most dramatic pain relief came from onabotulinumtoxinA injected into the suture lines alone (13). There is also recent documentation that a suture only onabotulinumtoxinA injection paradigm was effective in patients with chronic migraine (14). In addition, extracranial administration of onabotulinumtoxinA can suppress the response of rat meningeal trigeminal nociceptors to stimulation and the response was greatest if the onabotulinumtoxinA was injected only into the cranial sutures versus a cranial suture and muscle combination (15). Thus, one could expect a positive pain effect when utilizing cranial suture injections in the treatment of PCH with suture disruption.
Limitations
This is a small case series thus the positive results must be replicated in a much larger population utilizing a stricter injection criterion. As only four patients have been treated using our combination injection protocol this is a significant limitation affecting the generalizability of our findings. We injected various amounts of onabotulinumtoxinA into areas of palpable tenderness from our examination of the skull, thus not all patients received the same amount of onabotulinumtoxinA in the same locations. In addition, a key to our protocol was injecting into the cranial suture lines. Palpating the cranial suture lines is not an easy task for the headache specialist or any medical practitioner especially when multiple suture lines are involved including the more minute ones. Thus, how precise we were when injecting the sutures lines is unknown. In addition, does precision matter or could just ‘following the pain’ provide the same results? If we could be more exact in our injections over the involved cranial suture lines and if we injected the entire length of the suture rather than just the painful regions would our results be even better?
Conclusion
From our case series utilizing a novel onabotulinumtoxinA injection protocol we saw very positive results in our patients with persistent PCH with transected cranial suture lines. It appears that injection not only along the painful craniotomy scar but to the involved cranial suture lines provides positive efficacy and sustained benefit. One of our patients (Patient 1) had the chronic migraine PREEMPT protocol completed for her PCH prior to seeing us and had no benefit after multiple cycles suggesting that focusing on the scar and suture lines may be essential for PCH treatment effect. It also demonstrates the absolute need for examining and palpating the skull of the patient prior to treatment to define the cranial suture lines involved which may include suture lines outside of the craniotomy surgical field. One of our outlier patients had no pain over her craniotomy site but pain only along her suture lines outside of the craniotomy region and did very well with suture injection alone. Prior positive studies of onabotulinumtoxinA for PCH could have also been injected into or near the cranial sutures and this just was not documented by the authors, or they did not recognize the suture involvement. There have been no significant side effects of chronic onabotulinumtoxinA treatment in our patients as well as those previously reported. There is definitely long-term consistency with our treatment paradigm and one of our patients had her PCH essentially cured. Further study of onabotulinumtoxinA for persistent PCH is needed in clinical trial form. The trial would need to utilize a defined injection protocol and group patients by craniotomy location and cranial suture line disruption thus helping to define if specific craniotomy sites and/or specific cranial suture lines are more amenable to onabotulinumtoxinA than others. Based on location of injection (away from the forehead) a placebo arm could be readily utilized as there would be no obvious loss of facial wrinkles to help denote who did and did not receive onabotulinumtoxinA. Three injection treatment arms would be required including craniotomy scar alone, cranial suture lines alone and a combination of both.
There is no defined preventive treatment for persistent post craniotomy headache. OnabotulinumtoxinA has shown some promise in case reports and small case series, but no specified injection protocol has been established. Post craniotomy headache maybe an extracranial based head pain syndrome caused by injury to skull based and cranial suture based trigeminal sensory nociceptors. It appears that onabotulinumtoxinA injection not only along the painful craniotomy scar but to the involved cranial suture lines provides positive efficacy and sustained benefit for persistent post craniotomy headache.
Footnotes
Authorship
JR: Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. TR: Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. AB: Conceptualization, Investigation, Writing – original draft, Writing – review & editing.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
