Intra-arterial Thrombolysis in Late Pregnancy
To the Editor:
Johnson et al injected a limited amount of recombinant tissue plasminogen activator (r-TPA) in the middle cerebral artery of a 37-week pregnant woman with left acute ischemic stroke.1 The middle cerebral artery did not reopen. On the final angiogram, an efficient collateral circulation trough posterior cerebral artery, not seen on the preoperative angiogram, filled the peripheral branches of the occluded middle cerebral artery. The patient rapidly improved, and no adverse effects of the procedure were observed.
We congratulate the authors for rapid diagnosis, excellent therapeutic approach and good final results, but we deeply disagree with the interpretation of angiographic pictures. Figure 1 (preoperative) and Figure 2A (postoperative) express exactly the same pattern of collateral circulation toward peripheral middle cerebral artery, the sole difference being the phase, early and late arterial, respectively. In effect, some thin branches of middle cerebral artery filled by leptomeningeal collaterals are already seen in Figure 1. In other words, collateral circulation may be similar pre- and postoperatively. In Figure 2B the main branches of the anterior cerebral artery are still injected, whereas posterior cerebral artery is not. This is the typical angiographic pattern of a posterior cerebral artery filled both from anterior and posterior circulation: the pressure wave of the contrast injection in carotid artery partially fills the posterior cerebral artery. Once the pressure wave of the injection is past, contrast is rapidly washed out from fresh blood of posterior circulation. Visualization of the posterior cerebral artery depends on the pressure of the injection and on the position of the catheter. The alternative hypothesis suggested by Johnson et al of an asymptomatic thrombus occluding posterior communicans artery at its origin, dissolved by 15 mg of r-TPA injected in the middle cerebral artery, seems less realistic. Improvement may be explained by natural history of the middle cerebral artery occlusion, frequently going toward rapid improvement with limited basal ganglia infarctions,2 possibly helped by the systemic action of r-TPA. Anyway, the behavior of the authors was absolutely correct, because it was based on the clinical and not on the angiographic improvement. In similar cases a mechanical embolectomy may represent a less risky alternative.
Johnson DM, Kramer DC, Cohen E, Rochon M, Rosner M, Weinberger J. Thrombolytic therapy for acute stroke in late pregnancy with intra-arterial recombinant tissue plasminogen activator. Stroke. 2005; 36: 53–55.
Phan TG, Donnan GA, Wright PM, Reutens DC. A digital map of middle cerebral artery infarcts associated with middle cerebral artery trunk and branch occlusion. Stroke. 2005; 36: 986–991.
I want to thank Drs Bergui and Bradac for their interest in our article and their kind words regarding our management. I am pleased to report the patient and her child continue to do well.
The writers suggest that the patient’s clinical improvement may have been an example of the natural history of resolving middle cerebral artery (MCA) occlusion with residual basal ganglia infarction. They disagree with our interpretation that improved collateral circulation to the MCA territory accounted for the clinical improvement. These writers offer a valid criticism because the few images published do not clearly depict this change in circulation.
During the thrombolysis itself, we observed the first appearance of the posterior communicating artery (PCOM) and posterior cerebral artery (PCA), and, with that, we saw improved collateral circulation to the MCA via PCA-MCA leptomeningeal collaterals (Figures 1 and 2⇑). The prethrombolysis image (Figure 1) shows nonfilling of the PCOM, with nonfilling of the parietal and temporal branches of the MCA. The postthrombolysis image (Figure 2), at comparable angiographic phase, using similar injection parameters, shows the first filling of the PCA with concurrent improved filling of additional parietal portions of the MCA territory.
These more complete images demonstrate that the differences in MCA collateral filling result from the improved PCA circulation, not from the differences in angiographic phase as suggested by the writers. One can never know whether the treatments given result directly in patient improvement, assist the patient in the natural process of healing, or occur contemporaneously with, but coincident to, spontaneous healing. We can only hope our efforts did contribute to the good outcome for this patient and her child.