Competing interest statement
Conflict of interest: the authors declare no potential conflict of interest.
There is still controversy about occurrence of post-procedural headaches and coil-embolization of intracranial non-ruptured aneurysms. There are studies indicating a relief of headache after coil-embolization1 and others showing an increase of headache after the procedure, at least for a short time.2 However, there are no studies evaluating the severity of headaches of endovascular vs surgical treatment in ruptured aneurysms. In the present study we aimed to compare the incidence of post-procedural headaches as assessed by the visual analog scale (VAS) after surgical clipping or interventional coiling in patients with aneurysmal subarachnoid haemorrhage (aSAH) of WFNS grades 1-3.
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. For this type of study formal consent is not required.
Materials and Methods
For the retrospective study there was no approval needed as given by the Physicians Chamber of Nordrhein.
Inclusion criteria of our present retrospective analysis were: i) patients with aneurysmal subarachnoid haemorrhage suffering from WFNS grade 1-3; ii) extubation of the patient at maximum 12 hours after aneurysm occlusion; iii) full orientation regarding time and situation; iv) reliable answering to questions without any aphasic disturbances; v) treatment between September 1st 2015 and September 1st 2016. Patients with non-aneurysmal or high-grade SAH, comatose, confused or aphasic patients were excluded.
After confirmation of an aneurysmal SAH by native computed tomography and CT-angiography, patients were admitted to the neuro intensive care unit (NICU) and treated according to the AHA- and institutional guidelines.3 All patients with a GCS < 12 received an external ventricular drainage (EVD) but most of these patients had a WFNS score of > 3 and where excluded from the study. Digital subtraction angiography was performed for aneurysm visualization and confirmation. Aneurysm was repaired either by microsurgical or endovascular technique within 24 hours after admission. After the procedure, patients in both groups were treated in the same manner, were transferred to the ICU and – if possible – were extubated. Prophylactic treatment with dihydopyridine L-type calcium channel antagonists was started and all patients underwent serial perfusion computed tomography (PCT) during their stay on the NICU.
Standard pain medication was paracetamol (3x1 g), metamizole (3x1 g) and – only if required – piritramide (3 mg i.v.).
Headache was assessed by the visual analog scale (VAS; no pain: 0, to maximal pain: 10), which was the main study parameter. The pain scores were recorded for the initial headache at aneurysm rupture, 24 h (N=43), 21 days (N=43), 3 months (N=19) and 6 months (N=3) after surgery. For the 24 h time point patients were asked 3 times a day in fixed time points and the mean VAS score was calculated in order to minimize any immediate effect of analgesics which could have possibly given to the patient at the time point of assessment of the VAS. At the same time, dosage and specific analgesia was recorded. Other parameters were age, gender, aneurysm location, vasospasms, ischemia.
Cerebral vasospasms were defined as transcranial Doppler mean velocity of MCA > 120 cm/sec, a Lindegaard Index of >4.5 or clinical deterioration.
We have tested whether pain perception differs between patients who received clipping and those who received coiling therapy using a two-sided Mann-Whitney-Wilcoxon test. Due to lack of feedback we could not test the difference at follow-up of 6 months. All statistical analysis was performed using the R statistical computing package, R version 3.2.2 as released on 2015-08-14 (https://r-project.org/).
Descriptive statistics including mean and standard deviation (SD) were calculated for all continuous variables.
Totally, 43 patients out of 67 patients with aneurysmal subarachnoid haemorrhage fulfilled the inclusion criteria. The mean WFNS grade was 1.95 (surgical group 1.95±0.8; endovascular group 1.95±0.9). There was no statistical difference in the WFNS grades of the clipped or the coiled patient group. Also there was no difference in the fisher grades. Comorbodities of patients between the 2 groups was not significantly different either. The groups seem to be homogenous.
The mean age was 58.4±10.4 years in the surgically treated group and 56.8 ±9.6 years in the endovascular group. The female:male ratio was insignificantly different with a f: m ratio of 1.6:1 in the surgically clipped population and 3.2:1 in the endovascular coiled population.
As expected, the aneurysm location in the two groups differs significantly. Medial cerebral artery aneurysms (MCA) were treated exclusively by surgical clipping (N=10). 8 anterior communicans artery aneurysms (AcomA) were clipped and 6 were coiled, whereas 3 posterior communicans artery aneurysms (PcomA) were clipped and 5 endovascularly treated. One pericallosal artery aneurysm was treated by surgery and 3 were treated endovascular. All the aneurysms of the posterior circulation (N=6) were coiled.
In the surgically clipped group, the pain decreased from initially 9.9±0.29 before surgery to 1.09±1.96 24 h after surgery. In the endovascular group the pain decreased from initially 9.14±1.68) to 5.6±2.07 in the same time point. This difference of VAS score decrease of 8.8±2.03 points in the surgical group vs 3.38±2.67 points in the endovascular group immediately after treatment was highly statistical significant at a P<0.001 (Figure 1). Twenty-one days after aneurysm treatment the VAS score was 0.68±1.46 in the clipped group and 1.8±1.99 in the coiled group with a p value of <0.05. Three months after treatment the pain score was 0.11±0.33 in the surgical population and 0.9±1.91 in the endovascular population. The pain score at this time point was not significantly different between the two groups.
Subgroup analysis of posterior vs anterior circulation aneurysms provided the same results as the whole group since all posterior circulation aneurysms were coiled (Figure 2).
Another subgroup analysis of anterior circulation aneurysms only showed that patients with surgically treated aneurysms recovered significantly faster than endovascular treated patients (Figure 3). To exclude the possibility of bias caused be a hydro-cephalic state in the coiling group, which could lead to a higher level of headaches, we evaluated the number of patients with development of hydrocephalus. N=9 patients suffered of hydrocephalus. Four patients were in the surgically treated group and 5 in the endovascular group. The patient number of hydrocephalus was almost similar between the 2 groups and can be excluded as a confounder.
Analgesia as a confounder?
We calculated the dosages of analgesia in the first 24 h for every patient. Almost every patient in the endovascular group received more analgesics as in the surgical group.
We treated the patients with novaminsulfon, paracetamol, dipidolor boluses of 7.5 mg and Sufentanol i.v.
The surgically treated patients received a mean of 2.7 g novaminsulfon ±0.65 (median: 3) and the endovascular treated patients received 2.66±0.79 g novaminsulfon (median: 3). The difference in the dosage was not significant. For dipidolor the surgically treated patients received 10.7±5.9 mg (median 7.5 mg, 1 bolus) and the endovascular treated patients 21.87±11.28 mg (median 22.5 mg, 3 boluses). The endovascular treated patients received significantly more dipidolor (P<0.05). The dosage of sufentanol i.v. was in both groups similar with 366.6±265 µg (median: 340) in the surgical group and 386±337 µg (median: 380) in the endovascular group. Paracetamol was also not significantly different with 1.25±0.5 g (median: 1) in the surgical group and 1.71±0.95 g (median: 1) in the endovascular group.
There seems to be a benefit of clipping vs coiling concerning post-procedural headaches, which can be observed immediately after surgery, remains significant even after 3 weeks and subsides after 3 months. Especially with coil embolization there are a number of studies indicating an increase of headaches immediately after the endovascular intervention of incidental aneurysms, which seems to subside after 3 months.2 These results are in perfect accordance to our study concerning ruptured aneurysms. Another study shows a 55.6% incidence of newly developed headaches in patients after endovascular procedure.2 A possible mechanism of post-coiling headaches could be local thrombotic events and the dilation of the vessel wall which leads to endothelial inflammation.2,4,5 The intracranial arteries (especially the proximal portions) are highly innervated by sensory nerves that could be activated by these inflammatory processes as well as by structural changes of the vessel walls caused be embolization and coil packing.2,6-8 Coil packing seems to play a crucial role in the development of post-coiling headaches. A coil attenuation of more than 25% was identified as a risk factor for post-coiling headache.9
We did not observe residual headaches in surgically clipped patients. Application of a surgical clip does not apply any additional pressure to the aneurysm wall, neither causes any inflammation during thrombotic embolization and the absence of these phenomena could explain why we did observe such a significant difference in the headache cessation of the surgically treated group compared to the endovascular group. On the other hand more aneurysms in the clipped group were MCA aneurysms and there is a possibility that the innervation of the more distally located vessel is different than the vessels, which are closer to the circle of Willis. But even the AcomA-aneurysms, which were surgically treated, showed an immediate alleviation of headaches after the procedure. The prevalence of craniotomy related headaches which varies from 0-50% has to be kept in mind concerning the long term effects of aneurysm treatment in the clipped group.2,10,11 Nevertheless, the long term effects of surgery vs endovascular coil embolization in concern of post-procedural headaches should be a subject of further clinical study.
However, our present analysis is limited because of its retrospective character and the measure of the pain with the semi-objective VAS as well as the limited number of patients. The retrospective study can lead to imbalances in the 2 populations but the two groups were as homogenous as possible for a retrospective study.
Other kinds of pain, like temporomandibular joint pain, which can appear after surgery were not assessed in the study. The main reason of not including this complain in the study was that the first 3 weeks after the bleeding it was less of a concern to the patients and at 3 months, patients were not having any complain in this area.
To our knowledge this is the first study evaluating the immediate effects of post-procedural headaches in surgically clipped vs endovascular-coiled patients with ruptured aneurysms of WFNS grades 1-3. It could be shown that after clipping the headaches almost disappeared after 24 h and do not reappear during the follow up period of 3 months whereas after coil-embolization, headaches were still present in the first 3 weeks and disappeared after 3 months.