Reducing Cerebrospinal Fluid Sampling Frequency and Costs in Patients With Ventriculostomy for Aneurysmal Subarachnoid Hemorrhage: A Quality Improvement Initiative

BACKGROUND: We implemented a quality improvement project to transition from routine cerebrospinal fluid (CSF) sampling to indication-based sampling in aneurysmal subarachnoid hemorrhage (aSAH) patients with an external ventricular drain (EVD). METHODS: Forty-seven patients were assessed across 2 epochs: routine (n=22) and indication-based (n=25) CSF sampling. The primary outcome was the number of CSF samples, and secondary outcomes included cost reductions and ventriculostomy-associated infections. RESULTS: Patient characteristics were similar in the routine and indication-based sampling groups, as was the mean (SD) EVD duration (13.86 [5.28] days vs. 12.44 [4.78] days, respectively; P =0.936). One hundred eight CSF samples were collected during the quality improvement project; 81 in the routine sampling period and 27 in the indication-based sampling period. The median (interquartile range) CSF sampling rate reduced from 4 (3 to 4) per patient during routine sampling to 1 (0 to 2) during indication-based sampling (odds ratio: 0.19; 95% CI: 0.08-0.46; P <0.001), representing a 73% reduction in the number of samples after the transition to indication-based sampling. Each CSF sample cost $723, resulting in total sampling costs in the routine and indication-based sampling periods of $58,571 and $19,524, respectively. Therefore, the mean cost per patient was significantly higher in the routine sampling period than in the indication-based period ($2772 [$615] vs. $889 [$165], respectively; P =0.007). There were no ventriculostomy-associated infections in either period. CONCLUSION: Transitioning from routine to indication-based CSF sampling in aSAH patients with an EVD reduced sampling frequency and associated costs without increasing infection rates.