Remarkable Recovery After a Large Thalamic-Midbrain Intracerebral and Intraventricular Hemorrhage

PRACTICAL IMPLICATIONS

Coma with signs of midbrain injury (extensor posturing and dilated nonreactive pupils) after intracerebral and intraventricular hemorrhage may not indicate irreversible damage, and aggressive medical therapy for an adequate duration may be necessary to promote remarkable functional recovery.

In patients with intracerebral hemorrhage (ICH), low Glasgow coma scale (GCS) score, intraventricular hemorrhage (IVH), and infratentorial location are associated with high mortality¹ and minimal chance of meaningful functional recovery.² This report describes a patient with thalamic and midbrain ICH/IVH, who presented with coma, nonreactive dilated pupils, extensor posturing and yet had a good neurologic recovery at 30 days and functional independence at 1 year.

A 32-year-old woman with chronic hydrocephalus status after ventriculoperitoneal shunt (VPS) placement, presented for an elective revision of right VPS. Intraoperative course was complicated by choroid plexus injury. The VPS was removed, and a right frontal external ventricular drain (EVD) was placed. Initial head CT scan showed an 8 mL right thalamic ICH/IVH (modified Graeb score [mGS] = 11) involving posterior horn of the right lateral ventricle (figure 1). The patient's initial GCS score was 10, with eye opening to voice, intact brainstem reflexes, and pain localization. Two hours later, the patient declined with no eye opening, intact brainstem reflexes, and extensor posturing (GCS score = 4). A follow-up head CT showed hematoma expansion to 20 mL with extension into midbrain and worsening IVH (mGS = 22), now involving bilateral ventricles with casting of the third and fourth ventricles and obstructive hydrocephalus (figure 2). After ICH expansion, the patient's ICH score increased from 2 to 4.

Figure 1. Initial Head CT.

(A) Initial noncontrast head CT showing a small right thalamic intracerebral hemorrhage and right lateral ventricular hemorrhage (IVH) with the ventriculoperitoneal shunt tip in the right lateral ventricle. (B) Small IVH in 4th ventricle, but no hydrocephalus or mass effect. IVH = intraventricular hemorrhage.

Figure 2. Follow-up Head CT.

(A) Repeated noncontrast head CT showing 20 mL right thalamic ICH with massive IVH, casting of right lateral ventricle and extension into left lateral ventricle. (B) Rostral midbrain ICH, IVH with casting of 3rd ventricle and left temporal horn, obstructive hydrocephalus, and mass effect. (C) IVH with casting of 4th ventricle, obstructive hydrocephalus, and mass effect. ICH = intracerebral hemorrhage; IVH = intraventricular hemorrhage.

On postoperative day (POD) 3, patient developed elevated intracranial pressure (ICP) up to 32 mm Hg and was found to have no eye opening to pain, dilated nonreactive pupils, intact corneal and cough reflex, and extensor posturing to pain (GCS score = 4; FOUR score = 4). CT showed persistent ICH, IVH with obstructive hydrocephalus, and transtentorial herniation. The patient was already receiving hypertonic saline infusion to maintain a sodium target of 145–155 mEq/L. Serum sodium was increased to 160 mEq/L, after which no further hyperosmolar therapy was given. Propofol infusion was administered to control ICP. A second EVD was placed on the left side. After CSF drainage for 48 hours, the ICP was maintained below 20 mm Hg and propofol infusion was discontinued; however, patient continued to be comatose with nonreactive pupils and extensor posturing.

MRI of the brain showed no brainstem infarction or injury. Somatosensory evoked potentials (SSEPs) showed intact N20 responses bilaterally. EEG showed severe diffuse background slowing with loss of sleep architecture, with evidence of variability and reactivity characterized by intermittent high-amplitude delta activity during arousal or stimulation. Given the patient's young age, lack of indicators suggesting poor likelihood of awakening and family's wishes, aggressive medical care was pursued. On POD 13, the patient was noted to have reduced pupillary size and reactivity to light (FOUR score = 7). By POD 17, the patient was opening eyes and following commands in right hemibody. On POD 20, she was able to move her left hemibody. Bilateral EVDs were removed, and right VPS was replaced. At discharge, patient was alert, attentive, and verbally communicative, but bradyphrenic, with persistent left hemiparesis. At the 1-year follow-up, patient had no deficits except upgaze impairment and mild ataxia. Her 1-year modified Rankin score was 2.

Discussion

The most commonly used tool for prognosis after ICH is the ICH score.¹ It predicts 30-day mortality but is over simplistic with many limitations. It was not designed to predict the long-term functional outcomes. The ICH score does not incorporate the IVH volume, which also affects mortality and functional outcomes.³ It does not incorporate the exact ICH location which also affects the outcomes. Brainstem and thalamic locations are associated with higher mortality and increased disability.⁴ The ICH score does not account for hematoma expansion, which is an independent outcome predictor.⁵ Finally, it does not account for early withdrawal of life-sustaining measures. Over reliance on such scales leads to a self-fulfilling prophecy because of premature limitation of care in patients with poor predicted outcomes, further propagating mortality.⁶

Noninvasive tests, such as SSEP, have been used to predict recovery from coma with limited data in ICH.⁷ The absence of the SSEP in comatose patients with ICH was associated with 0%–4% chance of awakening.⁷

Our patient's ICH score was 4, with 97% predicted 30-day mortality. Despite this, the patient made a remarkable recovery with functional independence at 1 year. Various explanations for patient's recovery exist. The patient's young age may have contributed. The patient had an intraoperative ICH, where immediate interventions to impede the hemorrhage were instated. The patient had extensive IVH which was aggressively drained. Importantly, continued aggressive care likely contributed to her outstanding recovery.

In conclusion, early prognostication leads to premature withdrawal of life-sustaining measures. The above case argues against such practice, especially in young patients who might make remarkable recovery if given maximal individualized interventions with aggressive neurocritical care and time.

Appendix. Authors

Study Funding

No targeted funding reported.

Disclosure

The authors report no disclosures relevant to the manuscript. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.