Evaluation of the SOFA score as a tool to predict DCI-associated infarctions after spontaneous subarachnoid hemorrhage

Elena Kurz^1*Verena Fassl¹Carolin Brockmann²Alicia Schulze³Darius Kalasauskas¹Florian Ringel¹Axel Neulen¹

• ¹Department of Neurosurgery, University Medical Centre, Johannes Gutenberg University Mainz, Mainz, Germany
• ²Universitatsmedizin der Johannes Gutenberg-Universitat Mainz Klinik und Poliklinik fur Neuroradiologie, Mainz, Germany
• ³Universitatsmedizin der Johannes Gutenberg-Universitat Mainz Institut fur Medizinische Biometrie Epidemiologie und Informatik, Mainz, Germany

Spontaneous subarachnoid hemorrhage (SAH) accounts for approximately 5–10% of all strokes and has high rates of mortality and long-term disability (1). In addition to early brain injury, which occurs as a direct consequence of intracranial bleeding and the associated transient global cerebral ischemia, delayed cerebral ischemia (DCI) is an important cause of secondary brain injury. In many patients, the course is further complicated by peripheral organ dysfunctions. In particular, neurogenic cardiomyopathy and neurogenic pulmonary edema occur in varying degrees in a large subgroup of SAH patients, as well as systemic inflammation, which can occur as a consequence of treatment-associated infections. The pathophysiology of DCI is complex and not completely understood (2). In addition to cerebral pathophysiological effects such as vasospasms of large and small arteries, cortical spreading depressions, microthrombosis, cerebral inflammatory processes, and others, peripheral organ dysfunction has been associated with DCI in several studies (2–5). In particular, cardiac dysfunction—by reducing cerebral blood flow—and systemic inflammation—by triggering neuroimmunological effects—have been suggested to contribute to DCI (6–8). Oral nimodipine, recommended in the weeks after SAH, has been demonstrated to reduce the rate of DCI-associated infarctions (9). Nevertheless, DCI still occurs in up to 30% of SAH patients and can lead to cerebral infarctions with consecutive functional deficits and a poor neurological outcome. Therapeutic options include inducing hypertension and endovascular salvage therapies in refractory cases. However, it is crucial to start these therapies before the manifestation of infarctions. Since early diagnosis of DCI is crucial, intensive monitoring of SAH patients at risk for DCI is a central element of therapy during the first 2–3 weeks after SAH (10). It remains challenging to appropriately select patients at risk of DCI. It has been shown that the degree of EBI correlates with the risk of DCI. Accordingly, relatively high predictive values for predicting DCI have been shown for the Hunt & Hess (H&H) and the WFNS scores, both of which reflect the severity of EBI (11, 12). These scores do not consider other organ dysfunctions. A score depicting inflammatory processes and dysfunction of other organ systems, in addition to brain injury, might be superior to these established scores in predicting DCI. The Sequential Organ Failure Assessment (SOFA) score is routinely used in intensive care medicine to monitor organ dysfunction and is associated with mortality in sepsis (13–15). Other studies have shown that the score provides a holistic picture of inflammation and reflects organ dysfunction in critical illness (13, 14). The SOFA score is calculated by evaluating mean arterial blood pressure (MAP) and the need for vasopressors as measures of cardiac function, the Glasgow Coma Scale (GCS) as a measure of brain injury, the Horovitz quotient as a measure of the oxygenation capacity of the lung, and serum bilirubin, platelet counts, and creatinine levels (14, 16). Lambden et al. (14) reported that an increase of 2 points in the SOFA score was associated with a mortality rate of approximately 10% in ICU patients with infections, and it has been shown to have better prognostic accuracy for the prediction of mortality in case of critical illness than other routinely used scores. The SOFA score has also been correlated with mortality and outcome after SAH (17, 18). It remains unclear, however, whether the SOFA score could also be used to predict DCI and to select SAH patients at high risk of DCI and DCI-associated infarctions. Therefore, our study aimed to investigate whether the SOFA predicts the occurrence of DCI-associated infarctions and to compare it to the established H&H and WFNS scores. We further set out to analyze the ability of the different parameters contributing to the SOFA score to predict DCI-associated infarctions and to compose a simplified new score based on the most robust parameters.