Introduction

Decompressive craniectomy (DC) is a well-established neurosurgical intervention in patients suffering from malignant infarction of the middle cerebral artery (MCA). While prospective randomized trials showed a decrease in mortality and an improvement in functional recovery, mechanical thrombectomies (MT) were not part of the standard treatment regimen in stroke patients back then [12, 16, 17, 26, 27].

In recent years, multiple randomized controlled trials showed a benefit for endovascular treatment in terms of functional outcome at 90 days and underlined the importance of MT as a further effective tool in large vessel occlusions that limits the extent of ischemic brain damage [5, 6, 10, 15, 24]. A significant reduction in the number of malignant infarctions and mortality rates due to endovascular treatment was demonstrated by Fuhrer et al. [7].

It may be assumed that with the introduction of MT, the clinical characteristics of patients undergoing an additional DC have changed. Additionally, one may expect a reduction in the overall rate of decompressive craniectomies after implementation of MT as has been demonstrated recently [3, 4, 22, 25].

In this context, the aim of this study was to investigate the influence of nowadays regularly performed mechanical thrombectomies on patients undergoing DC.

With regard to their clinical course and outcome, we compared patients undergoing DC subsequently to mechanical thrombectomy to those patients in whom MT was not a component of the treatment.

Methods and materials

All patients undergoing DC due to malignant MCA infarction between January 2009 and January 2018 were included in this study. To investigate the influence of MT on the collective of patients with DC, only infarctions of thrombotic or embolic origin were analyzed. Patients who underwent MT prior to DC were compared to DC patients without endovascular treatment.

Clinical data

Patients’ demographics and clinical data, including sex, age, laboratory findings, side of DC, and the number of modifiable cardiovascular risk factors (CVRF) were collected retrospectively. CVRF included hypercholesterolemia, arterial hypertension, diabetes mellitus, obesity, nicotine and alcohol abuse, and chronic renal insufficiency. Additionally, initial National Institutes of Health Stroke Scale (NIHSS) and the time interval from symptom onset to DC were compared between patients with and without thrombectomy. Functional outcome was assessed via modified Rankin Scale (mRS) at the end of rehabilitation.

Imaging and treatment

The extent of the stroke prior to surgery was determined using the Alberta Stroke Program Early CT score (ASPECTS) on CT imaging (Fig. 1).

Fig. 1
figure 1

Collecting the Alberta Stroke Program Early CT Score (ASPECTS). 1 point is deducted from the initial score of 10 for every MCA vascular territory involved. Case courtesy of Dr. Subash Thapa, Radiopaedia.org, rID: 40018

Full size image

The indication for endovascular treatment was predominantly based on a multimodal CT examination: non-enhanced CT, CT angiography, and, in about half of the cases, an additional CT perfusion. In 11% of all patients, a magnetic resonance imaging was performed.

Endovascular treatment in our institution included access via an 8 F femoral sheath and an 8 F guiding catheter/balloon catheter for the carotid artery. An access via brachial, radial, or carotid puncture was not performed in this series. The degree of reperfusion achieved was quantified using the TICI-score which has been described by Higashida et al. [11].

Early decompressive craniectomy was performed in patients at risk of malignant cerebral swelling. Each indication was verified by an interdisciplinary team of the senior neurosurgeon and neurologist in charge, based on patient’s characteristics (age, medical history, laboratory findings), neurological condition, and neuroradiological imaging. This strategy corresponds with the current guideline of the German Society of Neurology, which recommends making the decision for or against DC on an individual basis [13].

A question mark–shaped skin incision was made fronto-parieto-temporal, ending approximately 10 mm anterior of the tragus, followed by retraction of the musculocutaneous flap, removal of the bone flap, and dural opening [14, 23].

Anticoagulants taken by patients before the event are discontinued perioperatively or, if possible, antagonized. A single thrombocyte aggregation inhibition might be maintained perioperatively, depending on the initial indication. In cases of dual antiplatelet therapy, the ADP receptor antagonist is paused.

Statistics

Statistical analyses were performed using IBM SPSS® v. 23 (IBM Corp., Armonk, NY, USA). Univariate analysis was performed using chi-squared tests or ANOVA tests, depending on the scale of the measurements, to examine correlations between the parameters. Probability values of p < 0.05 were considered statistically significant.

This study was conducted according to the Declaration of Helsinki, local and institutional laws, and was reported to the local ethical committee (No.: WF-09/19).

Results

The number of patients with ischemic strokes treated in our institution has increased annually in recent years. While 465 ischemic strokes were recorded in 2009, the admission rate increased to 731 in 2017 (Fig. 2). During this period, the absolute number of DCs remained constant at an average of 13 per year, while a relative decrease in the rate was recorded: in 2009, 2.8% of ischemic stroke patients underwent surgery, compared with 1.9% in 2017 (Fig. 2). During this period, mechanical thrombectomies were introduced and doubled in number within 3 years from n = 84 in 2014 to n = 160 in 2017 (see also Fig. 2).

Fig. 2
figure 2

Temporal development of ischemic infarction, thrombectomy, and DC

Full size image

In total 119 DCs due to malignant MCA infarction were performed between January 2009 and January 2018. Ninety-six patients (59 males; 61.5%) were eligible for inclusion due to thrombotic or embolic origin of the infarction. One third (n = 32) of these patients had undergone MT prior to surgery. TICI-scores from 0 to 3 were reported in this group, while successful reperfusion (TICI-score 2b/3) was achieved in 13 patients (40.7%, Table 1).

Table 1 TICI-scores in thrombectomized patients
Full size table

In comparison, DC patients without thrombectomy had no statistically significant differences with respect to their clinical characteristics: both groups did not differ in age, sex, or neurological status at admission including the presence of anisocoria (Table 2).

Table 2 Comparison of thrombectomized and non-thrombectomized patients
Full size table

Comparing thrombectomized and non-thrombectomized patients, the mean age was 54 vs. 54.7 years and the proportion of female patients was 37.5 and 39.1%, respectively. Median initial NIHSS was 18 in thrombectomized patients vs. 18 in non-thrombectomized patients. The mean number of modifiable cardiovascular risk factor was 1.5 vs. 1.7. No differences could be seen on laboratory findings at admission such as C-Reactive Protein (CRP) (16.2 and 28.4 mg/dl, respectively) or hemoglobin (8.6 vs. 8.4 mmol/l). Likewise, the groups did not differ regarding coagulation parameters such as international normalized ratio (INR) or partial thromboplastin time (PTT) (Table 2). The time between onset of symptoms and surgery did not differ significantly in the groups. In addition, there was no difference in the ASPECT score, describing the same extent of infarction on CT-imaging (median 0 in both groups).

No significant difference could be shown regarding the neurological outcome after completion of rehabilitation. The mRS was assessed 372 days after surgery (mean) and did not differ in the groups with median scores of 4 in both groups.

Discussion

The encouraging results of mechanical thrombectomies in stroke patients have changed the standard treatment regimen with a continuous increase of MTs during the last years. At our institution, we could record a parallel relative decline in the number of decompressive craniectomies. This decrease is most likely attributable to the increase in interventional treatments. This trend following the introduction of routine thrombectomies has also been described in other studies [3, 4, 22, 25].

In our cohort of MT patients that required DC, 40% of the patients had a successful MT with a TICI-score of 2b or more on final angiography (complete or nearly complete reperfusion). The fact that these patients required DC suggests that even successful MT does not exclude all patients from life-threatening clinical courses requiring decompressive surgery. The outcome of the MT depends on various factors: in addition to the timing and success of the revascularization, some patients also experience very early severe infarct edema, which is not reversible despite thrombectomy [20, 21]. Therefore, in some cases, an operation remains necessary even after a successful MT.

Patients who had undergone thrombectomy prior to surgery did not differ from those without MT regarding their clinical parameters. Elevated CRP and anemia at admission have been described as prognostic markers for worse outcome in infarct patients [9, 18]. Our patient groups did not differ in this respect. Likewise, no differences were found between the groups with respect to cardiovascular risk factors as indicators of the cardiovascular risk profile of patients [8, 19]. Radiologically, the ASPECT-score is reported to have a prognostic significance regarding the functional outcome [1, 2]. No differences between patient groups could be found regarding the extent of the infarct, quantified by ASPECT-score or the functional outcome. Furthermore, an endovascular treatment performed prior to craniectomy did not lead to a delay of the surgery as there were no differences between patient groups regarding the timing of the operation.

Although our study has some limitations in its retrospective and single-center character, the results support the thesis that present studies on decompressive craniectomy in malignant MCA infarction can be compared to previous data and that they are still valid in times of thrombectomy due to a comparable patient collective [12, 16, 17, 26, 27].

In conclusion, the introduction of routinely performed mechanical thrombectomy as part of the standard treatment regimen for ischemic stroke has led to a relative decrease in decompressive craniectomies over time. Nevertheless, a successful thrombectomy does not seem to be a guarantee that surgical decompressive craniectomy will not be required in some patients.

Since DC patients with and without MT showed no differences regarding their initial clinical criteria and outcome, the following conclusions can be drawn: on the one hand, it can be concluded that mechanical thrombectomy does not increase the risks of subsequently performed DC since the outcome in the groups did not differ. On the other hand, these results further suggest that earlier studies on DC in patients with malignant MCA infarction also apply today for the collective of mechanically thrombectomized patients as the collectives are comparable.