The subset of ischemic stroke can be divided into hyperacute, acute, subacute and chronic stroke based on timing from the onset of stroke symptoms. It is generally a definition of time which is the first 6 hours, 6-48 hours, 48h to weeks, and weeks to months respectively [ref1]. However such duration does not have general agreement among various articles [ref2][ref3][ref4][ref5]. Thus, the author recommends that readers should not be obsessed with the specific duration of time as it does not carry a significant clinical implication. This is because, the computer tomography (CT) scans finding on each stage varies among individual. It depends on the size, duration and severity of infarct, the metabolic state of brain tissues and the presence of collateral arteries.

The ischemic changes detected by unenhanced CT are caused by cytotoxic edema (occurred within 30 minutes) attributed to lactic acidosis and failure of cellular membrane’s ion pumps. This results in redistribution of water from extracellular to intracellular space. This eventually reduces the Hounsfield values (The degree of attenuation detected by CT is measured by Hounsfield scale in “HU”). For every 1% increase in tissue water content, there is a reduction of X-ray attenuation by 3-5% which is equivalent to a reduction of 2.5 HU on CT. Tissue water content is increased by 0.9% at 2.5 hour and 2% at 4 hour. These small changes are difficult to be detected by human naked eyes. [ref6][ref7]

Hyperacute and Acute Stage

The CT findings of ischemic stroke depend on the artery and site involved. The CT findings during the first 3-6 hour (hyperacute) include loss of gray-white matter differentiation of cortical gyrus, basal ganglia or insula; loss of cortical sulci or narrowing of the Sylvian fissure; compression of ventricular system and basal cisterns; area of hypodensity; and hyperdensity in a circle of Willis vessel [ref1][ref6][ref8][ref9][ref10][ref11]. However, within 3 hours of symptoms onset, these early ischemic changes are found only in 31-53% of the patient but improves at 6 h with sensitivity of 67% [ref6][ref12]. Although normal CT brain might be found, it does not excludes the presence of ischemic stroke as it has a low negative predictive value (27%) in the first 6 hour [ref12]. If middle cerebral artery (MCA) is involved, it has been stated that up to 75% of the cases will have abnormal CT findings within 3 hour [ref8]. However in cases of lacunar infarct, only approximately 50% shows abnormal CT findings within 48 hour [ref 12][ref13][ref14].

Image above shows the early CT sign (<6h) of ischemic stroke with “loss of gray-white matter differentiation” in basal ganglia (left) and a diffuse cerebral swelling with los of cortical sulci and compression of ventricular system (right). Image obtained from [ref2].

Image above shows the early CT sign (<6h) of ischemic stroke with hyperdensity of MCA representing an acute embolus lodged into it (left) known as “hyperdense MCA sign”. Image on the right shows the hypodensity of insular cortex known as “insular ribbon sign”. Image obtained from [ref2].

Image above is a CT of ischemic stroke at 24h after onset, showing a hypodense area and enlargement of left temporal lobe which compresses the left Sylvian fissure. Image obtained from [ref1].

Image above shows the CT finding of left hyperdense MCA sign, and a dot sign (right) presence at the left Sylvian fissure. Image obtained from [ref8].

Due to the fact the ischemic stroke is a dynamic process, the CT findings change over time. The early ischemic changes can be present anytime up to 72hour regardless of the initial presence of CT findings [ref1]. Hypodensity area detected by CT previously may gradually expand, involving both gray and white matter of the brain.

Image above is the contrast-enhanced CT image at 4h after stroke onset (left) and non-enhanced CT at 32h (right) of the same patient. This demonstrate the expansion of hypodensity area as time progresses. Image obtained from [ref11].

Subacute Stage

As time progresses, in the subacute phase, brain swelling and mass effect will gradually build up within a week followed by gradual improvement beginning from that 1 week onward. These are not easily picked up by human eyes on CT. Initial hypodensity detected by CT usually remains during this phase. However, an interesting phenomena sometimes occurred during this phase known as “CT fogging effect’ where hypodensed infarcted area disappear, becoming isodense. This is probably dues to resolution of edema in the infarcted area. This usually occurs between 2-6 weeks after the onset of stroke. Such “disappeared infarct” will reappear in later phase in a form of tissue cavitation (encephalomalacia). [ref1][ref2][ref3][ref6][ref16]

In addition to that, there is also a risk of hemorrhagic transformation in 15-20% of the cases during this period of time. Most of the time, this occurred within 4-6 days after onset of stroke. Once happened, the hyperdensity CT image may persist up to 8-10weeks. [ref1][ref6]

Image above shows the CT result demonstrating the “fogging effect” occurs during subacute phase. Left CT image is obtained at 36h with bilateral occipital hypodensities. Right image is taken at 18 days showing the isodense appearance of previous infarct. Image obtained from [ref6].

Image above shows the CT result demonstrating the hemorrhagic transformation with foci of hemorrhage at the right post central gyrus (right). Image obtained from [ref7].

Chronic Stage

In chronic stage, which has vaguely defined period (weeks to months), the damaged necrotic tissue is resorbed. This results in formation of encephalomalacia accompanied by gliosis of adjacent brain tissue. Associated with this is dilation of ventricular system of affected part, though usually found in relatively large infarct. These pathological finding could be picked up by non-enhanced CT and MRI. [ref1][ref2][ref3]

Image above shows the CT picture of various location of ischemic stroke at chronic phase, demonstrating the encephalomalacia. Image obtained from [ref1].

This post is dedicated to my friend, Kenneth Tee.

References

1. Valery N. Kornienko and Igor N. Pronin, eds. Diagnostic Neuroradiology. First edition. Springer, Leipzig, Germany. 2009.

2. Mara M. Kunst and Pamela W. Schaefer. Ischemic stroke. Radiol Clin N Am. 2011; 49:1-26. Doi: 10.1016/j.rcl.2010.07.010

3. D. Nagaraja and N. Karthik. Imaging in stroke. Medicine Update. 2011; 214-220.

4. K. Aho et al. Cerebrovascular disease in the community: results of a WHO collaborative study. Bull WHO. 1980;58:113–130

5. John R. Lynch et al. Novel Diagnostic Test for Acute Stroke. Stroke. 2004; 35: 57-63. Doi: 10.1161/​01.STR.0000105927.62344.4C

6. R.G Gonzalez et al. Acute ischemic stroke: imaging and intervention. Second edition. Springer, New York, NY. 2011.

7. David Vu and Michael H. Lev. Noncontrast CT in acute stroke. Semin Ultrasound CT MRI. 2005; 26:380-386.

8. Keith W. Muir et al. Imaging of acute stroke. Lancet Neurology. 2006; 5:755-768.

9. Joanna M. Wardlaw and Orell Mielke. Early signs of brain infarction at CT: observer reliability and outcome after thrombolytic treatment, systematic review. Radiology. 2005; 235:444-453

10. Gyanendra Kumar et al. Penumbra, the basis of neuroimaging in acute stroke treatment: current evidence. Journal of Neurological Sciences. 2010; 288:13-24.

11. James M. Provenzale et al. Assessment of the patient with hyperacute stroke: imaging and therapy. Radiology. 2003; 229:347-359.

12. Jose G. Merino and Steven Warach. Imaging of acute stroke. Nature Review Neurology. 2009; 6:560-571

13. C. Stapf et al. Predictive value of clinical lacunar syndromes for lacunar infarcts on magnetic resonance brain imaging. Acta Neurologica Scandinavica. 2000; 101(1): 12-18

14. B. Norrving and S. Cronqvist. Clinical and radiologic features of lacunar versus nonlacunar minor stroke. Stroke. 1989; 20:59-64

15. Kavian Ghandehari and Zahra Izadi. Clinical evaluation of 625 lacunar syndrome patients. Turk J Med Sci. 2009; 39(4):607-612.

16. H. R. Jager. Diagnosis of stroke with advanced CT and MR imaging. British medical bulletin. 2000; 56(2):318-333