Subarachnoid Hemorrhage

Intro to SAH

Normal MCA velocities are 30-80 cm/s. Only low (< 120 cm/s) or high (> 200 cm/s) absolute velocities consistently correlate with arterial narrowing [Neurosurgery 44: 1237, 1999] – since most patients fall in an intermediate range, invasive treatment should not be based on these values alone. Instead, TCDs should be used as a precursor to formal angiography.

The hypotensive effects of nimodipine may necessitate dosing 30 mg every two hours (as opposed to 60 mg every four hours).

Early studies suggested that volume expansion, independent of effects on CPP, improved cerebral blood flow [Acta Neurol Scand Suppl 60: 490, 1979]. More recent data, however, suggest that prevention of hypovolemia (and not institution of hypervolemia) is critical. [Stroke 31: 383, 2000]

Because cerebral vessels lose the ability to autoregulate after SAH, CBF is directly proportional to MAP [J Neurosurgery 76: 493, 1992; J Neurosurgery 26: 299, 1967; Neurosurgery 28: 41, 1991]. Not surprisingly, several studies have shown the ability of induced hypertension to reduce delayed ischemic deficits [Stroke 18: 365, 1987; J Neurosurgery 72: 559, 1990; J Neurosurgery 49: 525, 1978; Anesthesiology 69: 116, 1988; J Neurosurgery 27: 441, 1967; Surg Neurol 25: 317, 1986; Acta Neurochir 103: 18, 1990; Neurosurgery 23: 699, 1998]. Triple H therapy has thus become well-accepted, despite the fact that, according to Andrews, there is only one published controlled trial to support it [Neurosurgery 12: 658, 1983]. In fact, hypervolemic therapy has only been studied well in 146 patients, 32 in the Norwegian study and the remainder in the Columbia study – both suggests that there is no reduction in delayed ischemia and no improvement, but that complications were more common [Neurosurgery 49: 593, 2001; Stroke 31: 383, 2000]. Importantly, triple H has been associated with CHF, pulmonary edema, electrolyte disorders, and brain edema.

20% of SAH patients will develop hydrocephalus [J Neurosurgery 63: 355, 1985]. In addition, a large percentage of poor grade (Hunt & Hess grade 4 or 5) patients will develop elevated ICP despite normal ventricular size, and these patients will actually improve when treated for hydrocephalus (do not lower pressure to less than 15 mm Hg, however). Subacute hydrocephalus can develop within 1 week, however these patients should only be treated if they’re symptomatic. Still, 20% of patients surviving SAH will need a shunt. [J Neurosurgery 86: 462, 1997]

Seizures in SAH

As many as 25% of SAH patients will suffer from seizures immediately following the event [Neurosurgery 8: 417, 1981; Can J Neurol Sci 13: 229, 1986], and another 3% will develop late-onset seizures [Neurosurgery 37: 863, 1995; Acta Univ Palacki Olomuc Fac Med 137: 61, 1994]. Many physicians advocate prophylactic use of AEDs but randomized controlled trials have failed to demonstrate a statistically significant benefit [J Neurol Neurosurg Psych 55: 753, 1992; Acta Neurochir Suppl 50: 52, 1990]. In fact, a recent study of 527 SAH patients showed that prophylactic phenytoin was associated with worse cognitive outcomes at 1 and 3 months. [Stroke 36:583, 2005]. Prior to this study, short term use was recommended [Neurosurgery 37: 863, 1995]. A retrospective analysis of phenytoin administration (3 days vs. until discharge) at Brigham and Women’s Hospital showed that the short term seizure rate was 1.3 vs. 1.9% (3d), p = 0.603 and that the long-term seizure rate: 5.7 vs. 4.6% (3d) at average follow-up period of 6.7 and 5.4 months, respectively, p = 0.573 [Neurosurgery 60: 99, 2007]

Seizure Prophylaxis in SAH Patients

  • No difference between phenytoin for 3 days or until discharge [Neurosurgery 60: 99, 2007]
  • Increased “Phenytoin burden” = with worse neurologic outcome [Stroke 36:583, 2005]

Medical consequences of SAH

Hyponatremia occurs in up to 30% [Stroke 22: 1519, 1991; Ann Neurol 17: 137, 1985]. CSW and SIADH are common causes, and may be differentiated by volume status. CSW is treated with hydration (0.9 or 3.0% NaCl) and salt tabs. It is the most common cause of hyponatremia in these patients and may respond to fludorcortisone [Stroke 20: 1156, 1989; Clin Neurol Neurosurg 90: 209, 1988]. SIADH is treated with fluid restriction (< 1L) initially, followed by a combination of 3% NaCl/furosemide if severe Cardiac arrhythmias occur in up to 30% of SAH patient but are usually non-sustained. Cardiac enzymes are often elevated, and are correlated with an increased risk of cardiopulmonary complications, delayed cerebral ischemia, and death or poor functional outcome at discharge [Circulation 112: 2851, 2005]. Hypervolemic and/or hyperdynamic therapy can exacerbate myocardial ischemia in these patients [Neurosurgery 11: 337, 1982] Pulmonary edema, pneumonia, and pulmonary embolism are the most common pulmonary complications, and together account for 50% of medical deaths in this patient population [Crit Care Med 23: 1007, 1995] Significant DVT and/or PE complicate at least 2% of SAH patients [Act Neurochir 133: 141, 1995]. Andrews recommends TEDs, SCDs and lovenox on all of these patients. One randomized, controlled trial of 120 patients actually showed a neurologic benefit of adding 20 mg SQ lovenox to SAH patients [Clin Neurol Neurosurg 106: 97, 2004], although this was refuted by another randomized, controlled trial of 170 patients (40 mg SQ) [J Neurosurg 99: 953, 2003] The overall incidence of GI bleeding is 4% [J Neurosurg 73: 18, 36]