Admitting adult craniotomy patients without significant comorbidities, who are expected to have a short length of stay, to the neuroscience ward has been associated with shorter stays and a reduced total cost of admission—with no significant differences in postoperative clinical outcomes. This is according to a retrospective analysis published in Neurosurgery.
In their report, Won Kim (Department of Neurosurgery, Ronald Reagan UCLA Medical Center, David Geffen School of Medicine at UCLA, Los Angeles, USA) and colleagues state that the neurointensive care unit (NICU) has traditionally been the default recovery unit after elective craniotomies, but add: “Our data suggest that even a single-night stay in the NICU was associated with increased overall hospitalisation cost and length of stay, but with no difference in clinical outcome (return to ED [emergency department], return to OR [operating room], and readmission within 30 days).
“The biggest contributor to this difference was the NICU’s base cost. Our data imply that adult patients without significant comorbidities and with an expected short length of stay who undergo elective craniotomy for supratentorial lesions may be safely admitted to a neuroscience ward unit without significant additional risks to the patient, while shortening their length of stay and reducing hospitalisation cost. Early recognition of this patient group and establishment of alternative pathways to ICU [intensive care unit] admission might decrease ICU utilisation, and the cost of hospitalisation.”
The researchers initiated a cost-effectiveness analysis of their own institutional experience over a five-year period to compare patients who recovered in the NICU versus the neuroscience ward following an elective craniotomy for tumour resection—the first time an analysis of a large series of patients such as this has been reported. They included all elective supratentorial craniotomy patients aged 18 years or older, with a length of stay of seven days or less, who were admitted between March 2013 and April 2018.
A total of 209 patients were admitted to the neuroscience ward, while 340 were admitted to the NICU. Any patient characteristics that may have been potential confounders were mostly comparable between these two groups, Kim et al claim. Their analysis revealed the average length of stay in the neuroscience ward group was 3.046 days, compared to 3.586 days in the NICU group (p<0.001). There was no difference in the expected length of stay between the two groups (p=0.184), Kim and colleagues report. “In our multivariate analysis, postoperative stay in the NICU, even for one night, was independently associated with increased length of stay (p=0.015), but not with expected length of stay,” they state.
In addition, the average total cost of hospitalisation in the NICU group was $3,193.33 more per admission than the total cost observed in the neuroscience ward group ($32,496.41 vs. $29,303.08; p<0.001). The authors also note that there was no statistically significant difference in surgery costs between the two groups—indicating that differences in costs were incurred postoperatively. In their post hoc analysis, they add, the differences in cost between the neuroscience ward and the NICU, stratified by length of stay, were greatest when the length of stay was one day, and this effect diminished with increasing length of stay. In multivariate analyses, neuroscience ward admission was a significant independent predictor for total cost when length of stay was less than two, three, four, and five days—suggesting that, for stays shorter than six days in length, neuroscience ward admission becomes “a strong determinant of total cost irrespective of intrinsic patient factors”.
Kim et al add that an evaluation of clinical outcomes in the neuroscience ward and NICU groups found no statistically significant differences in rates of return to the operating room or emergency department, or readmission within 30 days. However, they also acknowledge several limitations of their study, including its retrospective nature, and a lack of randomisation to either group, multicentre data to reduce institutional bias, or more detailed data assessing patient comorbidities.
They conclude that, in patients with a length of stay shorter than five days, direct postoperative neuroscience ward admission may even be an independent predictor of reduced hospitalisation costs, without significant differences in clinical outcomes. However, commenting on the generalisability of this study in discussion with NeuroNews, Kim tempered this conclusion by stating: “It is truly excellent ward-level care that allows us to use our centre as a surgical step-down unit. As such, these findings may not be translatable to other institutions where this is not the case.”