There is growing evidence supporting the role of inflammation in early brain injury and cerebral vasospasm following subarachnoid hemorrhage (SAH). Matrix metalloproteinases (MMPs) are released by inflammatory cells and can mediate early brain injury via disruption of the extracellular matrix and mediate vasospasm by cleaving endothelin-1 into vasoactive fragments. We hypothesize that inflammation marked by neutrophil elevation and MMP-9 release in human SAH is associated with vasospasm and with poor clinical outcome. We enrolled consecutive SAH subjects (N = 55), banked serial blood and cerebrospinal fluid (CSF) samples, and evaluated their 3-month modified Rankin scores (mRS). Vasospasm was defined as >50% vessel caliber reduction on angiography 6–8 days post-SAH. A poor outcome was defined as mRS > 2. We compared blood leukocyte and neutrophil counts during post-SAH days 0–14 with respect to vasospasm and 3-month outcome. In a subset of SAH subjects (N = 35), we compared blood and CSF MMP-9 by enzyme-linked immunosorbent assay (ELISA) on post-SAH days 0–1, 2–3, 4–5, 6–8, and 10–14 with respect to vasospasm and to 3-month outcome. Persistent elevation of blood leukocyte (p = 0.0003) and neutrophil (p = 0.0002) counts during post-SAH days 0–14 are independently associated with vasospasm after adjustment for major confounders. In the same time period, blood neutrophil count (post-SAH days 2–3, p = 0.018), blood MMP-9 (post-SAH days 4–5, p = 0.045), and CSF MMP-9 (post-SAH days 2–3, p = 0.05) are associated with poor 3-month SAH clinical outcome. Neutrophil count correlates with blood MMP-9 (post-SAH days 6–8, R = 0.39; p = 0.055; post-SAH days 10–14, R = 0.79; p < 0.0001), and blood MMP-9 correlates with CSF MMP-9 (post-SAH days 4–5, R = 0.72; p = 0.0002). Elevation of CSF MMP-9 during post-SAH days 0–14 is associated with poor 3-month outcome (p = 0.0078). Neither CSF nor blood MMP-9 correlates with vasospasm. Early rise in blood neutrophil count and blood and CSF MMP-9 are associated with poor 3-month SAH clinical outcome. In blood, neutrophil count correlates with MMP-9 levels, suggesting that neutrophils may be an important source of blood MMP-9 early in SAH. Similarly, CSF and blood MMP-9 correlate positively early in the course of SAH, suggesting that blood may be an important source of CSF MMP-9. Blood and CSF MMP-9 are associated with clinical outcome but not with vasospasm, suggesting that MMP-9 may mediate brain injury independent of vasospasm in SAH. Future in vitro studies are needed to investigate the role of MMP-9 in SAH-related brain injury. Larger clinical studies are needed to validate blood and CSF MMP-9 as potential biomarkers for SAH outcome.

No abstract

Background Subarachnoid hemorrhage (SAH) pathophysiology involves neurovascular proteolysis and inflammation. How these two phenomena are related remains unclear. We hypothesize that matrix metalloproteinases (MMPs) mediate the depletion of anti-inflammatory plasma-type-gelsolin (pGSN). Methods We enrolled 42 consecutive SAH subjects and sampled CSF and blood on post-SAH days 2–3, 4–5, 6–7, and 10–14. Controls subjects were 20 consecutive non-SAH hydrocephalus patients with lumbar-drains. LISA, western blotting and zymography were used to quantify pGSN and MMP-9. Results In CSF, pGSN was lower in SAH compared with control subjects on post-SAH days 2–3 (p =0.0007), 4–5 (p=0.041), and 10–14 (p=0.007). In blood, pGSN decreased over time (p=0.001), and was lower in SAH compared to control subjects on post-SAH days 4–5 (p=0.037), 6–7 (p=0.006), and 10–14 (p=0.006). Western blots demonstrated that SAH CSF had novel bands at 52 and 46 kD, representing cleaved pGSN fragments. Gelatin zymography showed that CSF MMP-9 was elevated in SAH compared with controls. Higher CSF MMP-9 correlated with lower CSF pGSN on post-SAH day 7 (r= −0.38; p= 0.05). Conclusion SAH is associated with decreased CSF and blood pGSN and elevated CSF MMP-9. Novel cleaved pGSN fragments are present in CSF of SAH subjects, consistent with pGSN cleavage by MMPs. Since pGSN is known to inhibit inflammatory mediators, these findings suggest that MMPs may reduce pGSN and exacerbate inflammation after SAH. Further studies are warranted to investigate the mechanisms underlying MMP-pGSN signaling in SAH.

This study compares a deep learning interpretation of 23 echocardiographic parameters—including cardiac volumes, ejection fraction, and Doppler measurements—with three repeated measurements by core lab sonographers. The primary outcome metric, the individual equivalence coefficient (IEC), compares the disagreement between deep learning and human readers relative to the disagreement among human readers. The pre-determined non-inferiority criterion is 0.25 for the upper bound of the 95% confidence interval. Among 602 anonymised echocardiographic studies from 600 people (421 with heart failure, 179 controls, 69% women), the point estimates of IEC are all <0 and the upper bound of the 95% confidence intervals below 0.25, indicating that the disagreement between the deep learning and human measures is lower than the disagreement among three core lab readers. These results highlight the potential of deep learning algorithms to improve efficiency and reduce the costs of echocardiography.