Microvascular cerebral blood flow response to intrathecal nicardipine is associated with delayed cerebral ischemia

Abstract

One of the common complications of non-traumatic subarachnoid hemorrhage (SAH) is delayed cerebral ischemia (DCI). Intrathecal (IT) administration of nicardipine, a calcium channel blocker (CCB), upon detection of large-artery cerebral vasospasm holds promise as a treatment that reduces the incidence of DCI. In this observational study, we prospectively employed a non-invasive optical modality called diffuse correlation spectroscopy (DCS) to quantify the acute microvascular cerebral blood flow (CBF) response to IT nicardipine (up to 90 min) in 20 patients with medium-high grade non-traumatic SAH. On average, CBF increased significantly with time post-administration. However, the CBF response was heterogeneous across subjects. A latent class mixture model was able to classify 19 out of 20 patients into two distinct classes of CBF response: patients in Class 1 (n = 6) showed no significant change in CBF, while patients in Class 2 (n = 13) showed a pronounced increase in CBF in response to nicardipine. The incidence of DCI was 5 out of 6 in Class 1 and 1 out of 13 in Class 2 (p < 0.001). These results suggest that the acute (<90 min) DCS-measured CBF response to IT nicardipine is associated with intermediate-term (up to 3 weeks) development of DCI.

Keywords: blood flow; diffuse correlation spectroscopy; nicardipine; subarachnoid hemorrhage; vasospasm.

Figure 1

Diffuse correlation spectroscopy (DCS) experimental setup. (A) Pictorial representation of the DCS sensor on the scalp and the optical interrogation path for the source and detector geometry (pink shaded region), along with a photograph of the sensor (inset). (B) The DCS sensor was secured to the patient's forehead (patient image obtained with specific consent). (C) Representative CT scan showing typical placement of the DCS sensor and external ventricular drain (EVD).

Figure 2

Effect of IT nicardipine on systemic blood pressure and heart rate: Mean arterial pressure [MAP, (A)], heart rate [HR, (B)], change in MAP from pre-administration levels [ΔMAP, (C)], and change in HR from pre-administration levels [ΔHR, (D)] as a function of time post-administration of IT nicardipine. Thin gray lines in each subplot denote individual patient responses, and thick blue lines denote the mean and standard deviation across all patients. The black vertical line at time = 0 denotes the start of treatment.

Figure 3

Relative changes in cerebral blood flow (rCBF) as a function of time post-IT nicardipine administration. (A) Individual patient rCBF trajectories (gray) along with the linear mixed-effects model fit (black). (B) Two distinct latent classes of rCBF response were identified by the latent class mixture model: Class 1 (n = 6, in magenta) showed minimal change in rCBF post-nicardipine, whereas Class 2 (n = 13, in green) exhibited an increase in rCBF over the course of monitoring. The model was overly sensitive to one distinct trajectory (in blue); this outlier trajectory was excluded from analysis. Here, thin lines denote individual patient responses, while thick lines denote the model best fit for each response class. (C) Bar plot of the indicidence of delayed cerebral ischemia (DCI) by rCBF response class.

Figure 4

Relationship between resting-state and pretreatment micro- and macrovascular blood flow. Mean blood flow velocity in the middle cerebral artery [MCA, (A)] and anterior cerebral artery [ACA, (B)] measured by transcranial Doppler ultrasound (TCD) on the hemisphere of DCS monitoring vs. diffuse correlation spectroscopy (DCS)-measured blood flow index (BFI). TCD and DCS data were both acquired prior to IT nicardipine administration.