Search Results

You are looking at 1 - 7 of 7 items for

  • Author or Editor: Craig A. Williamson x
  • All content x
Clear All Modify Search
Full access

Rafael Wabl, Craig A. Williamson, Aditya S. Pandey, and Venkatakrishna Rajajee

OBJECTIVE

Data on long-term functional recovery (LFR) following severe brain injury are essential for counseling of surrogates and for appropriate timing of outcome assessment in clinical trials. Delayed functional recovery (DFR) beyond 3–6 months is well documented following severe traumatic brain injury (sTBI), but there are limited data on DFR following severe cerebrovascular brain injury. The objective of this study was to assess LFR and DFR in patients with sTBI and severe stroke dependent on tracheostomy and tube feeding at the time of discharge from the intensive care unit (ICU).

METHODS

The authors identified patients entered into their tracheostomy database 2008–2013 with sTBI and severe stroke, encompassing SAH, intracerebral hemorrhage (ICH), and acute ischemic stroke (AIS). Eligibility criteria included disease-specific indicators of severity, Glasgow Coma Scale score < 9 at time of tracheostomy, and need for tracheostomy and tube feeding at ICU discharge. Assessment was at 1–3 months, 6–12 months, 12–24 months, and 24–36 months after initial injury for presence of tracheostomy, ability to walk, and ability to perform basic activities of daily living (B-ADLs). Long-term functional recovery (LFR) was defined as recovery of the ability to walk or perform B-ADLs by the 24- to 36-month follow-up. Delayed functional recovery (DFR) was defined as progression in functional milestones between any 2 time points beyond the 1- to 3-month follow-up.

RESULTS

A total of 129 patients met the eligibility criteria. Functional outcomes were available for 129 (100%), 97 (75%), 83 (64%), and 80 (62%) patients, respectively, from assessments at 1–3, 6–12, 12–24 and 24–36 months; 33 (26%) died by 24–36 months. Fifty-nine (46%) regained the ability to walk and 48 (37%) performed B-ADLs at some point during their recovery. Among survivors who had not achieved the respective milestone at 1–3 months, 29/58 (50%) were able to walk and 28/74 (38%) performed B-ADLs at 6–12 months. Among survivors who had not achieved the respective milestone at 6–12 months, 5/16 (31%) were able to walk and 13/30 (43%) performed B-ADLs at 12–24 months. There was no significant difference in rates of LFR or DFR between patients with sTBI and those with severe stroke.

CONCLUSIONS

Among patients with severe brain injury requiring tracheostomy and tube feeding at ICU discharge, 46% regained the ability to walk and 37% performed B-ADLs 2–3 years after injury. DFR beyond 1–3 and 6–12 months was seen in over 30% of survivors, with no significant difference between sTBI and severe stroke.

Full access

Joseph R. Linzey, James F. Burke, Jeffrey L. Nadel, Craig A. Williamson, Luis E. Savastano, D. Andrew Wilkinson, and Aditya S. Pandey

OBJECTIVE

It is unknown what proportion of patients who undergo emergent neurosurgical procedures initiate comfort care (CC) measures shortly after the operation. The purpose of the present study was to analyze the proportion and predictive factors of patients who initiated CC measures within the same hospital admission after undergoing emergent neurosurgery.

METHODS

This retrospective cohort study included all adult patients who underwent emergent neurosurgical and endovascular procedures at a single center between 2009 and 2014. Primary and secondary outcomes were initiation of CC measures during the initial hospitalization and determination of predictive factors, respectively.

RESULTS

Of the 1295 operations, comfort care was initiated in 111 (8.6%) during the initial admission. On average, CC was initiated 9.3 ± 10.0 days postoperatively. One-third of the patients switched to CC within 3 days. In multivariate analysis, patients > 70 years of age were significantly more likely to undergo CC than those < 50 years (70–79 years, p = 0.004; > 80 years, p = 0.0001). Two-thirds of CC patients had been admitted with a cerebrovascular pathology (p < 0.001). Admission diagnosis of cerebrovascular pathology was a significant predictor of initiating CC (p < 0.0001). A high Hunt and Hess grade of IV or V in patients with subarachnoid hemorrhage was significantly associated with initiation of CC compared to a low grade (27.1% vs 2.9%, p < 0.001). Surgery starting between 15:01 and 06:59 hours had a 1.70 times greater odds of initiating CC compared to surgery between 07:00 and 15:00.

CONCLUSIONS

Initiation of CC after emergent neurosurgical and endovascular procedures is relatively common, particularly when an elderly patient presents with a cerebrovascular pathology after typical operating hours.

Full access

Christopher Roark, David Case, Mark Gritz, Patrick Hosokawa, David Kumpe, Joshua Seinfeld, Craig A. Williamson, and Anne M. Libby

OBJECTIVE

Aneurysmal subarachnoid hemorrhage (aSAH) has devastating consequences. The association between higher institutional volumes and improved outcomes for aSAH patients has been studied extensively. However, the literature exploring patterns of transfer in this context is sparse. Expansion of the endovascular workforce has raised concerns about the decentralization of care, reduced institutional volumes, and worsened patient outcomes. In this paper, the authors explored various patient and hospital factors associated with the transfer of aSAH patients by using a nationally representative database.

METHODS

The 2013 and 2014 years of the National Inpatient Sample (NIS) were used to define an observational cohort of patients with ruptured brain aneurysms. The initial search identified patients with SAH (ICD-9-CM 430). Those with concomitant codes suggesting trauma or other intracranial vascular abnormalities were excluded. Finally, the patients who had not undergone a subsequent procedure to repair an intracranial aneurysm were excluded. These criteria yielded a cohort of 4373 patients, 1379 of whom had undergone microsurgical clip ligation and 2994 of whom had undergone endovascular repair. The outcome of interest was transfer status, and the NIS data element TRAN_IN was used to define this state. Multiple explanatory variables were identified, including age, sex, primary payer, median household income by zip code, race, hospital size, hospital control, hospital teaching status, and hospital location. These variables were evaluated using descriptive statistics, bivariate correlation analysis, and multivariable logistic regression modeling to determine their relationship with transfer status.

RESULTS

Patients with aSAH who were treated in an urban teaching hospital had higher odds of being a transfer (OR 2.15, 95% CI 1.71–2.72) than the patients in urban nonteaching hospitals. White patients were more likely to be transfer patients than were any of the other racial groups (p < 0.0001). Moreover, patients who lived in the highest-income zip codes were less likely to be transferred than the patients in the lowest income quartile (OR 0.78, 95% CI 0.64–0.95). Repair type (clip vs coil) and primary payer were not associated with transfer status.

CONCLUSIONS

A relatively high percentage of patients with aSAH are transferred between acute care hospitals. Race and income were associated with transfer status. White patients are more likely to be transferred than other races. Patients from zip codes with the highest income transferred at lower rates than those from the lowest income quartile. Transfer patients were preferentially sent to urban teaching hospitals. The modality of aneurysm treatment was not associated with transfer status.

Full access

Lynze R. Franko, Kyle M. Sheehan, Christopher D. Roark, Jacob R. Joseph, James F. Burke, Venkatakrishna Rajajee, and Craig A. Williamson

OBJECTIVE

Subdural hematoma (SDH) is a common disease that is increasingly being managed nonoperatively. The all-cause readmission rate for SDH has not previously been described. This study seeks to describe the incidence of unexpected 30-day readmission in a cohort of patients admitted to an academic neurosurgical center. Additionally, the relationship between operative management, clinical outcome, and unexpected readmission is explored.

METHODS

This is an observational study of 200 consecutive adult patients with SDH admitted to the neurosurgical ICU of an academic medical center. Demographic information, clinical characteristics, and treatment strategies were compared between readmitted and nonreadmitted patients. Multivariable logistic regression, weighted by the inverse probability of receiving surgery using propensity scores, was used to evaluate the association between operative management and unexpected readmission.

RESULTS

Of 200 total patients, 18 (9%) died during hospitalization and were not included in the analysis. Overall, 48 patients (26%) were unexpectedly readmitted within 30 days. Sixteen patients (33.3%) underwent SDH evacuation during their readmission. Factors significantly associated with unexpected readmission were nonoperative management (72.9% vs 54.5%, p = 0.03) and female sex (50.0% vs 32.1%, p = 0.03). In logistic regression analysis weighted by the inverse probability of treatment and including likely confounders, surgical management was not associated with likelihood of a good outcome at hospital discharge, but was associated with significantly reduced odds of unexpected readmission (OR 0.19, 95% CI 0.08–0.49).

CONCLUSIONS

Over 25% of SDH patients admitted to an academic neurosurgical ICU were unexpectedly readmitted within 30 days. Nonoperative management does not affect outcome at hospital discharge but is significantly associated with readmission, even when accounting for the probability of treatment by propensity score weighted logistic regression. Additional research is needed to validate these results and to further characterize the impact of nonoperative management on long-term costs and clinical outcomes.

Restricted access

Badih J. Daou, Siri Sahib S. Khalsa, Sharath Kumar Anand, Craig A. Williamson, Noah S. Cutler, Bryan L. Aaron, Sudharsan Srinivasan, Venkatakrishna Rajajee, Kyle Sheehan, and Aditya S. Pandey

OBJECTIVE

Hydrocephalus and seizures greatly impact outcomes of patients with aneurysmal subarachnoid hemorrhage (aSAH); however, reliable tools to predict these outcomes are lacking. The authors used a volumetric quantitative analysis tool to evaluate the association of total aSAH volume with the outcomes of shunt-dependent hydrocephalus and seizures.

METHODS

Total hemorrhage volume following aneurysm rupture was retrospectively analyzed on presentation CT imaging using a custom semiautomated computer program developed in MATLAB that employs intensity-based k-means clustering to automatically separate blood voxels from other tissues. Volume data were added to a prospectively maintained aSAH database. The association of hemorrhage volume with shunted hydrocephalus and seizures was evaluated through logistic regression analysis and the diagnostic accuracy through analysis of the area under the receiver operating characteristic curve (AUC).

RESULTS

The study population comprised 288 consecutive patients with aSAH. The mean total hemorrhage volume was 74.9 ml. Thirty-eight patients (13.2%) developed seizures. The mean hemorrhage volume in patients who developed seizures was significantly higher than that in patients with no seizures (mean difference 17.3 ml, p = 0.01). In multivariate analysis, larger hemorrhage volume on initial CT scan and hemorrhage volume > 50 ml (OR 2.81, p = 0.047, 95% CI 1.03–7.80) were predictive of seizures. Forty-eight patients (17%) developed shunt-dependent hydrocephalus. The mean hemorrhage volume in patients who developed shunt-dependent hydrocephalus was significantly higher than that in patients who did not (mean difference 17.2 ml, p = 0.006). Larger hemorrhage volume and hemorrhage volume > 50 ml (OR 2.45, p = 0.03, 95% CI 1.08–5.54) were predictive of shunt-dependent hydrocephalus. Hemorrhage volume had adequate discrimination for the development of seizures (AUC 0.635) and shunted hydrocephalus (AUC 0.629).

CONCLUSIONS

Hemorrhage volume is an independent predictor of seizures and shunt-dependent hydrocephalus in patients with aSAH. Further evaluation of aSAH quantitative volumetric analysis may complement existing scales used in clinical practice and assist in patient prognostication and management.