Article Text

Download PDFPDF

Decompressive craniectomy after intra-arterial thrombolysis: safety and outcome
  1. Urs Fischer1,
  2. Philipp Taussky2,
  3. Jan Gralla3,
  4. Marcel Arnold1,
  5. Caspar Brekenfeld3,
  6. Michael Reinert2,
  7. Niklaus Meier1,
  8. Marie-Luise Mono1,
  9. Gerhard Schroth3,
  10. Heinrich P Mattle1,
  11. Krassen Nedeltchev1
  1. 1Department of Neurology, Inselspital, University Hospital Bern, University of Bern, Bern, Switzerland
  2. 2Department of Neurosurgery, Inselspital, University Hospital Bern, University of Bern, Bern, Switzerland
  3. 3Department of Neuroradiology, Inselspital, University Hospital Bern, University of Bern, Bern, Switzerland
  1. Correspondence to Dr Urs Fischer, Department of Neurology, University of Bern, Inselspital, Freiburgstrasse 4, 3010 Bern, Switzerland; urs.fischer{at}insel.ch

Abstract

Objectives Data on decompressive craniectomy (DC) after intra-arterial thrombolysis (IAT) for treatment of malignant middle cerebral artery (MCA) stroke are lacking.

Methods The authors compared consecutive patients with middle cerebral artery (MCA) strokes who underwent decompressive craniectomy (DC) after IAT with DC patients without prior thrombolysis.

Results Thirty of 2395 consecutive patients with acute ischaemic stroke or transient ischaemic attack were treated with DC because of space-occupying oedema 12–72 h after symptom onset. Fifteen patients underwent intra-arterial thrombolysis (IAT) prior to DC. Baseline characteristics did not differ between thrombolysed and non-thrombolysed patients except for lower National Institute of Health Stroke Scale score in non-thrombolysed patients (median National Institute of Health Stroke Scale 17 vs 14, p=0.033). The outcome at 3 months was favourable (modified Rankin scale 0–3) in seven (47%) thrombolysed patients and in four (27%) non-thrombolysed patients (p=0.45). Mortality and major complications after DC did not differ between the two groups (p>0.05). Older age (p=0.037) and previous hypertension (p=0.047) independently predicted unfavourable outcome in DC patients, but not IAT prior DC.

Conclusions There was no difference of outcome of patients with DC with or without prior IAT. DC after IAT did not cause any more complications than DC without prior thrombolysis.

  • Intra-arterial thrombolysis
  • decompressive craniectomy
  • outcome
  • cerebrovascular
  • interventional
  • neurosurgery
  • stroke

Statistics from Altmetric.com

A pooled analysis of three randomised controlled trials (RCTs) showed that decompressive craniectomy (DC) in patients with malignant middle cerebral artery (MCA) stroke reduces mortality and increases the chance of favourable outcome.1 However, most patients in these trials were not treated with thrombolysis shortly before DC.

There is a major concern that DC after thrombolysis may result in bleeding during and after surgery, since the coagulation system is impaired. It is therefore unknown whether DC is beneficial for stroke patients who fail thrombolysis and subsequently develop life-threatening space-occupying oedema.

In this study, we aimed to compare outcome of patients with malignant MCA strokes who underwent DC after IAT with DC patients without prior thrombolysis.

Patients and methods

From January 2000 to April 2009, 2395 patients were admitted to the department of neurology of the University Hospital in Bern for ischaemic stroke or transient ischaemic attack (TIA). Thirty patients developed space-occupying infarction of the MCA territory and were treated with DC. Fifteen of them underwent intra-arterial thrombolysis (IAT) prior to DC, while 15 did not receive any thrombolytic agents.

MCA strokes were considered malignant when they involved more than 50% of the MCA territory on CT or MR imaging with a corresponding neurological deficit and a decrease in consciousness.2

Patients admitted within 6 h after stroke onset were considered candidates for thrombolysis. IAT was performed if: (1) a diagnosis of ischaemic stroke was established; (2) the baseline National Institute of Health Stroke Scale (NIHSS) score was ≥4 points or isolated hemianopia or aphasia were present;2 (3) time of symptom onset was clearly defined; (4) treatment could be initiated within 6 h from symptom onset; and (5) the patient or their family consented to arteriography and potential thrombolysis.3 In addition, mechanical thrombus disruption or thrombaspiration was performed in five out of 15 thrombolysed DC patients.

Patients who did not meet the criteria for thrombolysis were given 250–500 mg of aspirin immediately after exclusion of intracranial haemorrhage by neuroimaging. Patients presenting within 8 h after symptom onset with contraindications for thrombolysis were considered candidates for mechanical thrombaspiration without administration of thrombolytic agents. Six of 15 non-thrombolysed DC patients underwent mechanical thrombaspiration.

DC was performed in patients with malignant MCA strokes mostly within 24 h and never later than 72 h, always with the consent of the patient or their family. DC consisted in a standard hemicraniectomy with duroplasty. In surviving patients, cranioplasty was performed after 3–6 months with the cryopreserved bone flap, acrylate, or patient-specific implant. Ancillary investigations were performed to assess the stroke aetiology according to the Trial of Org 10172 in Acute Stroke Treatment criteria.4

Symptomatic intracranial bleeding after decompression, worsening of pre-existing ICH or other life-threatening infections such as septicaemia were considered major complications. Asymptomatic haemorrhagic transformations, extracranial bleedings, fluid pads or non-life-threatening infections were considered minor complications. Symptomatic and asymptomatic bleedings were classified according to the PROACT II classification.

Outcome was assessed 3 months after stroke by a neurologist using the modified Rankin scale (mRS).5 mRS scores of 0–3 were considered ‘favourable,’ and mRS scores of 4–6 were considered ‘poor’ outcomes.

Differences between DC patients with and without prior thrombolysis were assessed using the Fisher exact test (for comparison of proportions), independent t test (for comparison of continuous variables) and Mann–Whitney U test (for comparison of ordinal variables). A two-sided p value <0.05 was considered statistically significant. Backward stepwise logistic regression analysis was used to assess predictors of favourable outcome in all DC patients.

Results

Baseline characteristics, rates of major and minor complications, and outcome of DC patients with and without prior thrombolysis are presented in table 1. Patient characteristics, vascular risk factors and stroke severity did not differ between thrombolysed and non-thrombolysed patients except for lower NIHSS score in non-thrombolysed patients (median NIHSS 17 vs 14, p=0.033). Patients in the thrombolysis group were more likely to have cardioembolic strokes and strokes of undetermined aetiology, whereas non-thrombolysed patients more often had carotid artery dissections. The rate of mechanical thrombaspiration and time from symptom onset to DC did not differ between the two groups (p>0.05).

Table 1

Baseline characteristics and outcome in patients with malignant middle cerebral artery infarction who underwent decompression surgery

Outcome at 3 months was favourable in seven patients (47%) of thrombolysed patients and in four (27%) of non-thrombolysed patients (p=0.45). In thrombolysed patients, seven had an mRS of 3, four had an mRS of 4, and four patients died. In the non-thrombolysed group, four patients had an mRS of 2, seven an mRS of 4, one had an mRS of 5, and three patients died. Mortality did not differ between the two groups (p>0.05). In the thrombolysis group, two patients died because of progressive swelling of the infarct and two because of ICH. Three patients in the non-thrombolysis group died because of progressive swelling of the infarct.

Major complications after decompression were similar in both groups (p>0.05). Three patients in the thrombolysed group had major complications: worsening of the pre-existing ICH (patient 3), de novo symptomatic ICH in the infarcted area (patient 6), and pulmonary embolism and septicaemia (patient 10). In the non-thrombolysed group, seven major complications occurred: one patient suffered a myocardial infarction in the postoperative period, four a life-threatening epidural haematoma and two a pneumonia with septicaemia. Minor bleeding complications after DC were similar in both groups (p>0.05).

In a multivariate analysis, older age (p=0.037) and previous hypertension (p=0.047) independently predicted unfavourable outcome but not thrombolysis prior to DC.

Discussion

This study analyses early DC in patients who developed malignant MCA strokes and compares patients with or without prior IAT. The outcome of patients with DC after IAT in this small series was similar to that of patients who did not receive IAT. DC after IAT did not cause any more complications than DC without prior thrombolysis.

There is a major concern that DC after thrombolysis may result in bleeding during or after surgery. Urokinase, which we used to perform IAT, decreases the levels of plasminogen and fibrinogen, and increases the level of plasmin and fibrin- and fibrinogen-splitting products, thus inhibiting coagulation. Changes in the coagulation system after local IAT for acute stroke persist up to 24–48 h, although the plasma half-life of urokinase is <20 min.6 Ueda et al showed that fibrinogen–fibrin degradation products and the plasmin–α2 plasmin inhibitor complex return to the pretreatment level within 24 h after intra-arterial administration of urokinase, but fibrinogen levels and α2–plasmin inhibitor activities were slightly decreased up to 48 h.

Two of the 15 patients thrombolysed before DC showed major intracranial bleedings and died. One of them (patient 3) bled into his malignant MCA infarct after IAT, and CT after DC showed a progression of the pre-existing ICH. Therefore, ICH either occurred spontaneously or was triggered by IAT but unlikely by DC. The other patient who bled (patient 6) was pretreated with aspirin and clopidogrel. DC was performed 42 h after symptom onset because of rapidly increasing intracranial pressure. The patient started bleeding severely during surgery and died on the following day. Her haemorrhage was most likely related to the dual antiplatelet treatment rather than to IAT, since the major effect of urokinase is expected to have subsided by this time. In the non-thrombolysed group, major bleeding complications occurred also. Four epidural haematomas were observed.

The best and safest time point to perform DC after IAT is unknown and cannot be derived from our data. After IAT with urokinase, coagulation may be severely impaired for 12–24 h. Thus, it might be wise to postpone DC to 24 h after thrombolysis, but the longer waiting might outweigh the benefit of DC as observed in RCTs.1

There are several limitations to the study. We compared DC patients with and without prior IAT. Patients were not randomised. Therefore, a selection bias is possible. However, the outcome of our DC patients with and without IAT is in the same range as that reported in the pooled analysis of RCTs,1 and DC was necessary because of failure of IAT. Furthermore, except for common routine tests of coagulation, we did not perform systematic laboratory studies to assess specific fibrinolytic/coagulation parameters before surgery. Monitoring of such parameters might be useful to reduce the risk of diffuse bleeding during surgery further in addition to letting time elapse prior to surgery. Finally, the sample size of this study is small, but it is a first report of patients who have undergone DC after IAT.

In conclusion, DC after IAT for treatment of malignant MCA stroke seems to be safe. The outcome of patients with DC in this small series was similar to the outcome of patients, who did not receive IAT prior to DC.

Acknowledgments

We thank P Ballinari, for statistical advice, and K Kraus, for editorial assistance.

References

Footnotes

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Ethics approval was provided by the University of Bern.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.