Background

Numerous interventions have been studied to decrease the risk of delayed cerebral ischemia (DCI) after aneurysmal subarachnoid hemorrhage (aSAH). Among these, targeted fluid administration to prevent hypovolemia has become one of the mainstays of treatment, although there is not clear consensus on fluid types and dosing. Human Albumin (HA) has been shown to have benefits in a number of other pathologies, including shock and trauma and some data points to benefits in preventing DCI for patients with aSAH. This study attempted to determine if HA administration decreased rates of DCI and to assess for adverse outcomes associated with HA.

Methods

This study was a non-randomized, quasi-experiment comparing patients in the comparison cohort 2015-2018 to historical controls admitted between 2011-2014 at a tertiary referral hospital.  Inclusion criteria were all adult patients with evidence of subarachnoid hemorrhage on head computed tomography (CT) or lumbar puncture. Patients were excluded if they died within 24 hours of presentation, patients with severe SAH, defined as World Federation of Neurosurgical Societies (WFNS) 4-5, or those who did not have aneurysm securement because of transition to palliative care.  Standard of care related to imaging, early aneurysm securement, nimodipine and TCDs was the same across both cohorts.

Patients in both groups received vascular imaging, early aneurysm treatment, invasive hemodynamic monitoring, nimodipine treatment, and vasospasm monitoring with Transcranial Dopplers (TCDs). The control group received intravenous crystalloid fluid administration at the discretion of the intensivist but was guided by pulmonary artery catheter monitoring in most high-grade SAH or hemodynamically unstable patients. Those in the intervention arm were treated with 5% intravenous Albumin infusion at a fixed dose of 60 g/day for the first 5 days of admission, as well as an intravascular volume goal of euvolemia. For the intervention group, this was explicitly determined by invasive hemodynamic monitoring utilizing PiCCO transpulmonary monitoring and global end-diastolic volume with a goal of euvolemia.

Outcomes

The primary outcomes were the incidence of DCI and death during ICU and hospital stay, and secondary endpoints included Glasgow Outcome Scale Extended (GOSE) score and death at 6 months. Adverse events including pulmonary edema, hypo- or hypernatremia, aneurysm re-bleeding, and vasospasm were also investigated. 

Analysis

Effects of the intervention on the primary and secondary outcomes were reported in terms of hazard ratios and relative risks, with multivariable logistic regression and sensitivity analyses performed to adjust for confounding variables while calculating the effects of the protocol on the intervention arm. Post hoc analyses were performed to adjust for competing risks, as DCI incidence could be biased because patients who experience DCI are more likely to die. Daily and cumulative volumes of fluids administered were also compared between groups to assess overall volume status.

Results

A total of 189 patients with a median age of 58 and of whom 72.2% were women were included in the final analysis, including 63 in the historical control arm and 126 in the intervention arm. The two groups were overall well-matched in terms of demographics, although the intervention group was slightly older and had a higher number of high-grade subarachnoid hemorrhages (defined as WFNS 4-5; 19% in historical controls and 34.9% in intervention group). For the primary outcome, there was a consistently lower incidence of DCI in the intervention group in the primary, adjusted, and post-hoc analyses. However, this did not translate to a lower incidence of ICU or hospital mortality, nor was there a significant difference in functional outcome at 6 months between groups.

In regard to secondary outcomes, those in the intervention arm had a significantly decreased risk of hyponatremia while also receiving less crystalloid fluid resuscitation during the time of albumin infusion. There was a non-significant increase in incidence of pulmonary edema in the intervention group (RR 2.7, 95% CI 0.8-8.8), but the study was not powered to detect a meaningful statistical difference in this metric between the two groups. There was no difference between the two groups for any of the other secondary outcomes, including vasospasm, re-bleeding, hydrocephalus, and pneumonia.

Commentary

The primary finding of this study is that for patients with aSAH, treatment with an early intervention of iso-oncotic albumin was associated with a decreased incidence of DCI without an increase in hospital complications. Albumin administration in the post-subarachnoid hemorrhage period has been studied in pre-clinical animal trials as well as prior retrospective studies, with findings that show multifactorial effects, ranging from decreasing inflammation, increasing neuroplasticity, promoting endothelial homeostasis, and augmenting hemodynamic stabilization. This study does have several limitations, specifically a small sample size, a lack of randomization, and fluid management in the control group that was not standardized and was left to the discretion of individual intensivists.

However, part of the explanation for the findings of this particular study could simply be the favorable impact of a more sophisticated invasive hemodynamic monitoring system for assessment of fluid status as well as more protocolized treatment of fluids and volume assessments in the intervention group. This type of monitoring has been independently evaluated in other subarachnoid hemorrhage studies with positive results, and therefore could represent a confounding variable in this study. That being said, the utility of albumin in the subarachnoid population is clearly one that has generated significant interest, and further randomized clinical studies are needed to address this question in more detail.