To the Editor: New-onset headache within a week after craniotomy is commonly recognized as acute post-craniotomy headache (PCH). This headache pain mainly occurs within the first few post-operative days and is remarkably associated with a younger age, the female sex, and the retrosigmoid approach.[1] The mechanisms of PCH mainly involve mechanical and chemical irritation and aseptic inflammation of the scalp muscles, periosteum of the skull, dura mater, and trigeminal nerves.[2] The frequent onset of PCH leads to impaired daily activities, decreased quality of life, prolonged post-operative recovery, and worsened social functioning. The differential diagnosis of acute PCH may also lead to the patient receiving many unnecessary tests, including invasive tests such as a lumbar puncture.

Cervicogenic headache (CEH) is defined as a headache attributed to disorders of the cervical spine and its component bony, disc, and/or soft tissue elements, with or without accompanying neck pain.[3] An earlier review documented that 3 of 95 patients who had undergone acoustic neuroma surgery developed CEH-like PCH.[4] The International Classification of Headache Disorders (ICHD)-3 criteria classify such disorders as a CEH caused by inappropriate positioning during the operation and should be precisely differentiated from PCH. The elimination of cervical pain sources remains a crucial component of the treatment of CEH, which is not included in the usual management of PCH. Undifferentiated CEH results in neglected management of cervical pain sources and risks of extensive analgesic consumption. However, exhaustive characteristics of CEH following acoustic neuroma resection and other craniotomies have yet to be disclosed. In this prospective study, we investigated the incidence, clinical manifestations, and risk factors of CEH following craniotomy.

This monocentric, prospective, observational study was designed based on an ongoing institutional PAINFREE program (2020–2023), which was proposed to improve post-craniotomy analgesia. The program protocol was reviewed and approved by the Institutional Ethical Review Board (Approval No. KY2020-008-02-1) of Beijing Tiantan Hospital. Consecutive patients admitted for elective craniotomy were screened. Those who signed the inform consent and agreed to participate in the study process were enrolled in the PAINFREE program. Participants who entered the program were asked about their overall post-operative pain symptoms by the program assistants until discharge. Subjects with positive compliance were therefore screened and interviewed by pain management specialists through consultations. Through the program, participants’ demographic, sociodemographic, admissive, peri-operative, and consultative data and information were collected and recorded by the program assistants and pain management specialists. Between Januaury 2020 and May 2022, consecutive adult patients enrolled in the PAINLESS program were included in the current study. Participants who developed life-threatening peri-operative complications, were admitted to the intensive care unit, or were uncooperative with the consultation inquiries or management plans were excluded.

Eligible participants were allocated to two groups based on whether new-onset CEH was diagnosed after craniotomy. Patients with suspected clinical manifestations were evaluated for CEH using the revised Cervicogenic Headache International Study Group and ICHD-3 criteria.[3,5] Manual therapy was denied for peri-operative safety considerations. Diagnosed patients were administered oral analgesics. Diagnostic nerve blockade (DNB) was applied if the headache pain was extremely severe or drug-resistant for both diagnostic and therapeutic purposes. The DNB was accomplished using individually formulated mixtures of 2% lidocaine or 1% ropivacaine and normal saline. Those who obtained a positive DNB were therefore confirmed to have a CEH. Other clinically suspected patients who showed little responsiveness to the DNB were diagnosed with headache following craniotomy attributed to extensive origination.

Data acquisition was completed by two of our program assistants. Patients’ demographic information (age, sex, and body mass index [BMI]), sociodemographic information (history of smoking and alcohol drinking), pre-existing comorbidities (history of cervicospinal disorder, chronic headache, psychiatric disorder, and systemic diseases), and peri-operative data (craniotomy etiology, surgical approach, incision length, operation duration, operating positioning, and anesthesia technique) were extracted from the PAINLESS case report forms. Information and data were obtained during the consultation (primary consultative diagnosis and its approaches, responsiveness to management approaches determined during the consultation, headache pain onset time, laterality, property, intensity [documented as 0 to 10 points on a numerical rating scale, in which 0 represented for no pain and ten denoted the greatest pain imaginable, and pain scores of 0 to 3 points denoted mild pain, 4 to 7 points denoted moderate pain, and 8 to 10 points denoted severe pain], provocation, and accompanying symptoms) using the PAINLESS case report form and the institutional case record system. Subjects were deidentified after the data acquisition and interpretation were completed according to the PAINLESS program protocol.

The statistical analysis was performed using R Studio software (RStudio v. 2021.09.0 + 351 “Ghost Orchid”; R Core Team, Vienna, Austria) and associated packages. Descriptive data are presented as the means ± standard deviations or median (interquartile range [IQR]) for continuous variables and n (percentages) for categorical variables. Univariate comparisons between two arms were performed using the chi-squared test or Fisher's exact test for categorical variables. Continuous variables were compared using Student's t test or the Wilcoxon rank sum test. We performed univariate and multivariate logistic regression analyses with CEH as a binary outcome variable to assess the associations with demographic, clinical, and peri-operative variables. The magnitude of the association with the clinical outcome was reported as the odds ratio (OR) (95% confidence interval [CI]). All tests were two-sided. A P value <0.05 was set to be statistically significant.

During the study period, 1629 consecutive patients cumulatively underwent elective craniotomy in the study setting [Supplementary Figure 1, https://links.lww.com/CM9/B420]. Two hundred patients were excluded based on the exclusion criteria (20 due to severe peri-operative complications, 157 due to intensive care unit admission, and 23 due to a lack of cooperation with the consultation or DNB), resulting in a total of 1429 patients enrolled in this prospective study. The mean age was 48.98 ± 13.40 years, and 53.4% (763/1429) were women. Headache following craniotomy was reported in 53.3% (761/1429) subjects, among whom 83 were clinically diagnosed with CEH through consultations. Conservative treatments were initially applied, and positive pain relief was obtained in 61.4% (51/83) patients. DNB was used as needed in the other 38.6% (32/83) individuals, among whom 96.9% (31/32) acquired immediate headache pain relief and were confirmed to have a CEH. One patient presented with a CEH that failed the anesthetic blockade. Further mental examinations indicated that this patient was under significant stress, and the diagnosis of CEH was denied. In summary, 82 patients were finally diagnosed with CEH based on clinical manifestations and interventional techniques. The incidence was 5.8% (82/1429). The incidence of pre-existing chronic headache (medical history, not a new-onset or acute exacerbation) was 9.8% (8/82) in the CEH population and 12.2% (165/1347) among participants without CEH. The between-group difference was 0.603 according to the Chi-squared test. Patients with CEH more frequently had a history of cervicospinal disorders (72.0% [59/82] vs. 19.5% [263/1347], P < 0.001), and the difference between groups was statistically significant. No differences were detected in age, sex, BMI, smoking, intemperance, and other pre-existing medical history. Among the peri-operative data and information, patients with CEH had longer operation time (6.45 [IQR: 5.43–7.38] hours vs. 5.10 [IQR: 4.60–5.70] hours, P < 0.001). Other data involving the craniotomy etiology, surgical approach, incision length, operation positioning, and anesthesia technique did not differ between patients with or without CEH following craniotomy [Supplementary Table 1, https://links.lww.com/CM9/B420].

Generally, CEH was present 6.0 [IQR: 5.0–8.0] days post-operatively. Headache pain was primarily described as unilateral (71/82, 86.6%), moderate to severe (65/82, 79.3%), dull (54/82, 65.9%), or drilling (21/82, 25.6%) headache pain. Headache was mostly provoked by neck movement or pressing (68/82, 82.9%). Reduced neck motion was reported in 93.9% (77/82) of patients. Nausea with or without vomiting was reported by 31.7% (26/82) of patients. In addition, concurrent neck pain, photophobia, and phonophobia were reported in 29.3% (24/82), 13.4% (11/82), and 6.1% (5/82) of patients, respectively [Supplementary Table 2, https://links.lww.com/CM9/B420]. All collected demographic, sociomedical, and peri-operative data and information were entered into the univariate logistic analyses. Continuous variables such as age, BMI, incision lengths, and operation duration were categorically transformed during the process. A history of cervicospinal disorders (OR: 10.57 [95% CI: 7.01–16.29], P < 0.001) and an operative duration >5 h (OR: 4.95 [95% CI: 2.91–9.07], P < 0.001) were suggestive of CEH following craniotomy. In multivariate analyses, the association remained statistically significant; a history of cervicospinal disorders (OR: 10.84 [95% CI: 7.15–16.80], P < 0.001) and an operation duration >5 h (OR: 5.18, [95% CI: 3.01–9.60], P < 0.001) were positively associated with the development of CEH following craniotomy [Supplementary Table 3, https://links.lww.com/CM9/B420].

In the study, we identified a set of characteristics and several variables associated with CEH following craniotomy. Pre-existing cervicospinal disorders and prolonged operation durations >5 h were associated with an increased risk of developing CEH following craniotomy. Previous studies demonstrated the incidence of CEH was 4.1% (75/1838) of the general population.[6] In our study, 5.74% (82/1429) of the included patients developed CEH after craniotomy, which was similar to and slightly higher than those of a previous investigation, in which the occurrence of CEH similar to PCH was 3.2%.[4] Regarding the onset time, CEH developed on the 6th (IQR: 5–8th) post-operative day, which is the late onset and did not match empirical recognition. We speculate that the late onset is due to the early post-operative administration of systemic corticosteroids the following neurosurgery. CEH is characterized by moderate to severe headaches in the occipitofrontal region that may be precipitated by head movement. The pain presents as uni/bilateral, not throbbing or lancinating in nature, with fluctuating severity in non-clustering episodes and variable duration, and is sometimes accompanied by ipsilateral neck, shoulder, or arm pain. Our results were consistent with the aforementioned findings. More than one-quarter of the CEH population failed to determine the characteristics of their headache. The authors proposed that concurrent post-craniotomy pain may interfere with CEH pain and lead to an ambiguous headache pain configuration. Provocative factors were found in 86.6% (71/82) of participants in the study. The most typical trigger was neck movement or pressing, which generated headaches in >80% (68/82) of participants. Other accompanying mechanical or autonomic symptoms reported in the current study were also consistent with previous investigations.[6,7]

Our study had limitations. Due to the observational nature, we failed to collect extensive imaging data. For the same reason, comparisons among CEH following craniotomy, acute PCH, post-craniotomy pain, and headache following craniotomy attributed to other causes were beyond our estimation. Additionally, data from the excluded participants were not included in our estimation in this analysis. These results warrant further studies.

In summary, CEH is an uncommon post-operative complication of elective craniotomy. Operations longer than 5 hours and pre-existing cervicospinal disorders were significantly associated with the development of CEH following craniotomy. We proposed that the findings would enable pain physicians and neurosurgeons to recognize CEH following craniotomy. As CEH and PCH may overlap within the first week post-operatively, the ICHD-3 criteria might be subjective. Our results should be carefully considered.

Conflicts of interest

None.

References 1. Ryzenman JM, Pensak ML, Tew JM Jr. Headache: a quality of life analysis in a cohort of 1657 patients undergoing acoustic neuroma surgery, results from the acoustic neuroma association. Laryngoscope 2005; 115:703–711. doi: 10.1097/01.mlg.0000161331.83224.c5. 2. Rocha-Filho PA. Post-craniotomy headache: a clinical view with a focus on the persistent form. Headache 2015; 55:733–738. doi: 10.1111/head.12563. 3. No authors listed. Headache classification committee of the international headache society (IHS) the international classification of headache disorders, 3rd edition. Cephalalgia 2018; 38:1-211. doi: 10.1177/0333102417738202. 4. Schankin CJ, Gall C, Straube A. Headache syndromes after acoustic neuroma surgery and their implications for quality of life. Cephalalgia 2009; 29:760–771. doi: 10.1111/j.1468-2982.2008.01790.x. 5. Antonaci F, Ghirmai S, Bono G, Sandrini G, Nappi G. Cervicogenic headache: evaluation of the original diagnostic criteria. Cephalalgia 2001; 21:573–583. doi: 10.1046/j.0333-1024.2001.00207.x. 6. Sjaastad O. Cervicogenic headache: comparison with migraine without aura; Vågå study. Cephalalgia 2008; 28: (Suppl 1): 18–20. doi: 10.1111/j.1468-2982.2008.01610.x. 7. Bogduk N, Govind J. Cervicogenic headache: an assessment of the evidence on clinical diagnosis, invasive tests, and treatment. Lancet Neurol 2009; 8:959–968. doi: 10.1016/s1474-4422(09)70209-1.