Sheng-Jean Huang

Decompressive craniectomy as the primary surgical intervention for hemorrhagic contusion

The standard surgical treatment of hemorrhagic cerebral contusion is craniotomy with evacuation of the focal lesion. We assessed the safety and feasibility of performing decompressive craniectomy and duraplasty as the primary surgical intervention in this group of patients. Fifty-four consecutive patients with Glasgow Coma Scale (GCS) scores of less than or equal to 8, a frontal or temporal hemorrhagic contusion greater than 20 cm(3) in volume, and a midline shift of at least 5 mm or cisternal compression on computer tomography (CT) scan were studied. Sixteen (29.7%) underwent traditional craniotomy with hematoma evacuation, and 38 (70.4%) underwent craniectomy as the primary surgical treatment. Mortality, reoperation rate, Glasgow Outcome Scale-Extended (GOSE) scores, and length of stay in both the acute care and rehabilitation phase were compared between these two groups. Mortality (13.2% vs. 25.0%) and reoperation rate (7.9% vs. 37.5%) were lower in the craniectomy group, whereas the length of stay in both the acute care setting and the rehabilitation phase were similar between these two groups. The craniectomy group also had better GOSE score (5.55 vs. 3.56) at 6 months. Decompressive craniectomy is safe and effective as the primary surgical intervention for treatment of hemorrhagic contusion. This study also suggests that patient with hemorrhagic contusion can possibly have better outcome after craniectomy than other subgroup of patients with severe traumatic brain injury.

Predictors of survival and functional outcome in acute stroke patients admitted to the stroke intensive care unit.

Multivariate models have not been widely used to predict the outcome of acute stroke patients admitted to the intensive care unit (ICU). The purpose of this study was to determine potential measures observed in the first 12 h post-stroke that predict early mortality and functional outcomes in ICU-admitted stroke patients. Eight hundred and fifty acute stroke patients (ischemic stroke, 508; intracerebral hemorrhage, 342) were included in this analysis between November 2002 and December 2006. Measures of interest were obtained in the first 12 h after onset of stroke were analyzed for three types of outcome: 3-month mortality, 3-month mortality or institutional care, and poor functional outcomes at discharge. Poor functional outcomes were defined as a Barthel index <80 or a Rankin scale >2. Multivariate regression models were used to determine the predictive value of the observed measures. After 3 months, 17% of patients had died; 21% were alive but being cared for in institutional settings; and 62% were alive and living at home. Functional status at discharge indicated 16% of patients had died, poor function in 50%, and good function in 34% of patients. Initial stroke severity, measured by National Institute of Health Stroke Scale, and dependence on a ventilator predicts 3-month mortality and poor outcome in all stroke patients. In addition, old age, previous stroke, and total anterior circulatory infarct were associated with poor outcome in ischemic stroke patients; old age, low body mass index and the presence of intraventricular hemorrhage were associated with poor outcomes in intracerebral hemorrhage patients. In conclusion, early stroke mortality and outcome at discharge can be predicted in the first few hours following an acute stroke for moderate to severe ICU-admitted stroke patients.

Third ventricle midline shift due to spontaneous supratentorial intracerebral hemorrhage evaluated by transcranial color-coded sonography

Objective. We aimed to assess the clinical usefulness of the third ventricle midline shift (MLS) evaluated by transcranial color‐coded sonography (TCCS) in acute spontaneous supratentorial intracerebral hemorrhage (ICH). Methods. Consecutive patients with acute (<24 hours after symptom onset) ICH were recruited for this TCCS study. Sonographic measurement of MLS and the pulsatility index (PI) of the middle cerebral arteries were compared with head computed tomographic (CT) data, including MLS, and hematoma volume. Poor functional outcome at 30 days after stroke onset was defined as modified Rankin scale greater than 2. Results. There were 51 patients with spontaneous supratentorial ICH who received CT and TCCS studies within a 12‐hour window. Correlation between MLS by TCCS (mean ± SD, 3.2 ± 2.6 mm) and CT (3.0 ± 2.4 mm) was high (γ = 0.91; P < .01). There was also a good linear correlation between hematoma volume and MLS by TCCS (γ = 0.81; P < .01). Compared with ICH volume less than 25 mL, those with greater volume had more severe MLS and a higher PI of the ipsilateral middle cerebral artery (P < .001). Midline shift by TCCS was more sensitive and specific than the PI in detecting large ICH (accuracy = 0.82 if MLS ≥ 2.5 mm), and it was also a significant predictor of poor outcome (odds ratio, 2.09 by 1‐mm increase; 95% confidence interval, 1.06–4.13). Conclusions. Midline shift may be measured reliably by TCCS in spontaneous supratentorial ICH. Our study also showed that MLS on TCCS is a useful and convenient method to identify patients with large ICH and hematoma expansion and to predict short‐term functional outcome.

Dynamic cerebral autoregulation in carotid stenosis before and after carotid stenting

BACKGROUND Impaired dynamic cerebral autoregulation (DCA) has been shown in patients with severe (> or =70%) internal carotid artery (ICA) stenosis, but DCA in moderate (50% to 69%) ICA stenosis, especially its response to carotid revascularization, has rarely been reported. Our study aimed to characterize DCA in severe and moderate ICA stenosis before and after carotid stenting. METHODS This study included 21 patients with ICA stenosis > or =50% who received carotid stenting. Data of arterial blood pressure and cerebral blood flow velocity of the middle cerebral artery, measured by transcranial Doppler, were collected for 10 minutes < or =24 hours before and after stenting. The DCA index, represented as aMx, was assessed by calculating the Pearson product-moment correlation coefficient of spontaneous arterial blood pressure and cerebral blood flow velocity fluctuations. The relationship between aMx and stenotic severity and also alternations of aMx before and after stenting were assessed. RESULTS Carotid stenting was effective to improve the DCA in the stenting side but not in the contralateral nonstenting side. In considering individual ICAs, the average aMx (mean +/- SD) increased significantly from ICA stenosis <50% (0.117 +/- 0.091) to 50% to 69% (0.349 +/- 0.144), 70% to 99% (0.456 +/- 0.147), and total occlusion (0.557 +/- 0.210; P < .05, P < .01, and P < .01, compared with 50% to 69%, 70% to 99%, or total occlusion with <50% stenosis). The correlation between the degree of ICA stenosis and the aMx was also significant (r = 0.693, P < .005). The aMx improved significantly in the stented side after carotid stenting in both moderate and severe ICA stenosis, and this finding was not affected by age, sex, risk factors, or clinical symptoms. CONCLUSIONS In addition to patients with severe carotid stenosis, patients with moderate carotid stenosis may also have impaired DCA that can be restored after carotid stenting.

Intracranial pressure model in intensive care unit using a simple recurrent neural network through time

Abstract This paper aims to establish a patients intracranial pressure (ICP) model in neurosurgical intensive care unit (NICU) using neural network. Non-invasive physiological signals from patients including mean arterial pressure (MAP), heart rate (HR), end-tidal of carbon dioxide (EtCO2) and regional cerebral oxygenation (rSO2) were measured. However, ICP remains ill-defined, complicated and non-linear because it is affected by many predictable and unpredictable factors. Our study employs the structure of recurrent network to develop a modified neural network algorithm called a simple recurrent neural network through time (SRNNTT). The proposed recurrent neural network combines Elman architecture of the simple recurrent network structure and backpropagation through time. In order to demonstrate the performance of the proposed model, four kinds of neural-network classifiers have been tested on Mackey–Glass differential-delay equation which is a chaotic time series signal. Finally, we used this SRNNTT model to build the ICP model using data from six head-injured patients. Although the accuracy of the ICP model is still far from ideal, the methodology used non-invasive vital signs (i.e., MAP, HR, EtCO2, and rSO2) to predict an invasive, dangerous and expensive signal (i.e., ICP) has achieved this monitoring system more safely and flexibly in NICU.

Clinical outcome of severe head injury in different protocol-driven therapies

In the past 5 years cerebral perfusion pressure (CPP) management has become mainstream in the treatment of severe head injuries. The American Association of Neurological Surgeons guidelines (2000) suggest that CPP should be maintained at least 70 mmHg; however, there is still debate about optimal CPP level. The purpose of this study is to evaluate the effectiveness of three widely used therapies: intracranial pressure (ICP)-targeted therapy, CPP-targeted therapy with CPP > 70 mmHg, and modified CPP-targeted therapy with CPP > 60 mmHg. The clinical procedures, complications, and patient outcomes are compared. Data including patient age, sex, initial Glasgow Coma Score (GCS), ICP, CPP, fluid status, amount of mannitol and vasopressor used, daily intake and output, complications, and clinical results were collected from 213 patients with severe head injuries over a 12-year period. Patients were categorized into three groups (ICP, CPP, modified CPP [mCPP]) according to treatment protocol used. Retrospective data collection was by chart review. The mortality rate was 28.6%, 14.3%, and 13.5% in groups ICP, CPP, and mCPP, respectively. Highest intake/output ratio, amount of vasopressor used, and pulmonary complication rates were seen in group CPP patients. Group mCPP patients showed the best clinical outcome and lowest complication rate. Though CPP-targeted therapy is the most recommended therapeutic protocol, our data showed that the outcome is as good in the mCPP-targeted group with CPP > 60 mmHg as in the CPP-targeted group, but complications are fewer in the mCPP group.

Survey of do-not-resuscitate orders in surgical intensive care units.

BACKGROUND/PURPOSE End-of-life decisions are always difficult and complex, especially in the surgical setting. This study examines the epidemiology of do-not-resuscitate (DNR) orders, and the clinical factors influencing DNR consent. The impact of DNR on treatment and resource use in the surgical intensive-care unit (ICU) is also assessed. METHODS This retrospective observational study was performed at National Taiwan University Hospital, a tertiary medical center in Taipei. A total of 14,698 patients were admitted to the surgical ICUs between January 2003 and December 2006. Of these, 13,825 (94.1%) survived to ICU discharge and 873 (5.9%) died. Of those that died, 278 (1.9% of total patients) went home to die due to terminal stage illness and 595 (4.0 % of total patients) died in the ICU. All mortality patients were included in this study. RESULTS Yearly DNR rates were all above 65%. The average interval from ICU admission to DNR consent remained stable at 11-13 days, but the interval from DNR consent to death increased over the study period, from 2.0 to 3.5 days. Discussion over DNR was mainly initiated by intensivists. Multivariate logistic regression analysis found that older age (odds ratio, 1.010; p = 0.017) was significantly associated with DNR consent. DNR patients had longer ICU stays, lower fraction of inspired oxygen, and less inotropic infusion, dialysis, transfusion, laboratory examination, and chest radiography, but more use of sedative drugs, analgesics, and nutrition support at the time of death. After DNR, the use of advanced antibiotics, chest radiography, laboratory examination, and transfusion decreased. Inotropic infusion, however, continued to significantly increase. CONCLUSION Although DNR was common in our surgical ICU patients, this request was signed late in the ICU course, when therapeutic options had been exhausted. Early initiation of DNR discussion should be promoted to improve end-of-life care and reduce futile treatments in the ICU.

Comparison of the Applicability of Rule-Based and Self-Organizing Fuzzy Logic Controllers for Sedation Control of Intracranial Pressure Pattern in a Neurosurgical Intensive Care Unit

This paper assesses the controller performance of a self-organizing fuzzy logic controller (SOFLC) in comparison with a routine clinical rule-base controller (RBC) for sedation control of intracranial pressure (ICP) pattern. Eleven patients with severe head injury undergoing different neurosurgeries in a neurosurgical intensive care unit (NICU) were divided into two groups. In all cases the sedation control periods lasted 1 h and assessments of propofol infusion rates were made at a frequency of once per 30 s. In the control group of 10 cases selected from 5 patients, a RBC was used, and in the experimental group of 10 cases selected from 6 patients, a self-organizing fuzzy logic controller was used. A SOFLC was derived from a fuzzy logic controller and allowed to generate new rules via self-learning beyond the initial fuzzy rule-base obtained from experts (i.e., neurosurgeons). The performance of the controllers was analyzed using the ICP pattern of sedation for 1 h of control. The results show that a SOFLC can provide a more stable ICP pattern by administering more propofol and changing the rate of delivery more often when rule-base modifications have been considered

Longitudinal outcomes of patients with traumatic brain injury: a preliminary study.

Primary objective: To examine the long-term global clinical outcomes of TBI patients from 1 week to 10 years post-injury. Research design: A retrospective study was conducted at a level I trauma centre. Methods and procedures: A total of 327 TBI patients were recruited retrospectively in this longitudinal study and the score of Glasgow Outcome Scale Extended (GOSE) was then documented as an indicator of the clinical outcomes. Main outcomes and results: The results presented that less than one-in-three patients recovered well within 2 weeks after trauma. Around 40% of patients were evaluated as favourable outcomes within 6 months post-injury. Less than 30% of TBI patients could gain favourable outcomes at 3 years post-injury and over half of patients still could not recover well at 6 years. Five patients’ outcomes were continuously followed until 10 years and the results indicated that they started to have good outcomes after 3 years post-injury. Conclusions: This study might be the first one to longitudinally evaluate the outcomes of patients from 1 week to 10 years post-injury and revealed that patients still may have difficulties in social interactions and family relationships until 6 years post-injury, even though they could live and work independently. Hence, professionals should pay attention to the underlying factors causing their difficulties within the long-term recovery process.

Withdrawal from long-term hemodialysis in patients with end-stage renal disease in Taiwan

Withdrawal from dialysis is ethically appropriate for some patients with multiple comorbidities and a shortened life expectancy. Taiwan has the highest prevalence of dialysis patients in the world, and the National Health Insurance (NHI) program offers renal replacement therapy free of charge. In this review, we discuss its current status and many background issues related to withdrawing dialysis from patients with advanced renal failure in Taiwan. Compared with dialysis therapy, the medical resources for hospice care are relatively sparse. Since the announcement of the Statute for Palliative Care in 2000, there has been a gradual improvement in the laws and health polices supporting dialysis withdrawal. Culture and social customs also have a significant impact on the practice of hospice care. Based on current evidence and in accordance with the local environment, we propose recommendations for the clinical practice of dialysis withdrawal and hospice care. There remains a need to expand upon the community-based hospice care and home care systems to better serve patients. In conclusion, there are cross-cultural differences relating to dialysis withdrawal between Taiwan and Western countries. Our experience and clinical recommendations may be helpful for the countries with NHI systems or for the Eastern countries.