American Heart Association

Monthly Archives: October 2014

Effects of Extracranial Carotid Stenosis on Intracranial Blood Flow

Rajbeer Singh Sangha, MD

Shakur SF, Hrbac T, Alaraj A, Du X, Aletich VA, Charbel FT, and Amin-Hanjani S. Effects of Extracranial Carotid Stenosis on Intracranial Blood Flow. Stroke. 2014

A higher degree of extracranial carotid stenosis is associated with increased stroke risk and has become a key determinant in treatment decision-making. The dominant mechanism via which strokes occur is thought to be thrombo-embolic, however it has been postulated that hemodynamically-consequential narrowing of the vessel lumen can also result in cerebral hypoperfusion and may even potentiate the effects of distal embolization. Shakur et al. looked to characterize impact of degree of stenosis, stenosis length, and residual lumen on intracranial blood flow in patients with extracranial carotid stenosis.



The study was a retrospective analysis of 105 patients that were identified having ≥ 50% carotid stenosis who underwent revascularization. Patients in this study had undergone quantitative flow measurements of the extracranial and intracranial arteries using quantitative magnetic resonance angiography (QMRA). On multivariate analysis, MCA flow ratio was not significantly associated with percentage stenosis, stenosis length, or residual lumen. However, mean MCA flow ratio was significantly lower in symptomatic compared to asymptomatic patients (0.92 vs. 1.22, P=0.001). In contrast, mean ICA flow ratio was similar among these two groups (0.55 vs. 0.55, P=0.99).
The study findings suggest that in symptomatic extracranial carotid disease, the reduction in MCA flow may play an important role, thus implicating intracranial hemodynamics in the pathophysiology of this disease. It would be interesting to be able to classify the level of collaterals in these patients as this may be a determinant in whether patients remain asymptomatic. Furthermore, more studies should be conducted to better quantify the characteristics present in patients who suffer from symptomatic extracranial carotid disease vs asymptomatic extracranial carotid disease. Elucidating the pathophysiological mechanisms will better allow us to stratify ischemic stroke risk in the asymptomatic population.  
By |October 17th, 2014|pathogenesis|0 Comments
Read More

Lefties are Never Right: Is Atherosclerotic plaque in the Left carotid artery more vulnerable than on the Right?

Michelle Christina Johansen, MD

Selwaness M, van den Bouwhuijsen Q, van Onkelen RS, Hofman A, Franco OH, van der Lugt A, et al. Atherosclerotic Plaque in the Left Carotid Artery Is More Vulnerable Than in the Right. Stroke. 2014

Left hemispheric strokes can be devastating, impacting the patient’s language center and leading to hemiparesis and hemiplegia. Large vessel atherosclerosis is an established stroke subtype and plaque located at the carotid bifurcation has been implicated in as high as 18% of all strokes. Selwaness et al open their paper by noting that a significantly higher proportion of ischemic events are diagnosed in the left hemisphere compared to the right.  The team hypothesize that the higher incidence of events occurring in the lefthemisphere is related to either a higher prevalence, severity or vulnerability of atherosclerotic disease in the left carotid artery. 




Carotid MRI’s were performed on 1414 stroke free participants to assess not only the location but also degree of stenosis and components of the carotid plaque. The authors invited participants from The Rotterdam Study, a prospective population based cohort study who were routinely undergoing carotid ultrasound to also undergo MRI imaging of the bilateral carotids. The mean age of the final cohort was 72 and 53% were male.  Image quality was considered sufficient in 95% of scans. Luminal stenosis was calculated using the NASCET criteria.  The investigators classified the composition of plaques as either lipid-rich, containing intraplaque hemorrhage or calcification based on imaging characteristics. 

Overall, 1196 subjects or 85% had plaque in both carotid arteries meaning only 218 subjects had unilateral plaques. Within these patients, the investigators found that left sided plaques were twice as prevalent as right sided with no sex predominance but those with unilateral left sided plaques tended to be younger (68 vs 71). The degree of luminal stenosis did not differ between right and left and clinically relevant stenosis defined by NASCET also did not differ. When the components were analyzed individually, lipid-rich plaques was slightly more prevalent on the left (27.6% vs 23.4% p 0.006) and intraplaque hemorrhage (IPH) was also more frequent in left carotid artery plaques (23.1% vs 19.7% p 0.01).  Calcification was equal on both sides. When a single or predominant component was assigned, IPH and lipid were most prevalent in left-sided plaques but this time right sided plaques were predominantly composed of calcification.

The conclusion of the investigators is that carotid atherosclerotic plaque size and composition are not symmetrically distributed and that plaques on the left are more vulnerable than on the right due to the presence of IPH versus calcification. 

This inference should give the practicing vascular neurologist pause. In treatment of asymptomatic carotid stenosis, the mantra has been best medical management. Many are familiar with the CREST data which showed that among asymptomatic patients, the primary outcome (periprocedural stroke, death and myocardial infarction rates) did not differ significantly between stenting and surgery (4.9% vs. 5.6%) but the study was not powered to obtain significance (p=0.07). There are ongoing trails (CREST-2) to evaluate if optimal medical management is in fact sufficient in these patients. If in fact left sided plaques are more vulnerable to rupture, while we wait for the outcome of clinical trials should a practicing Neurologist change practice? For example, would one lower the blood pressure of a patient with left sided stenosis more aggressively than their right sided counterpart? Is dual antiplatelet therapy warranted even in the absence of intracranial stenosis? The authors appropriately discuss many limitations in evaluation of their data to include the fact that significance was only obtained when the plaques were assigned a predominant component, a small n and potential observer bias. This limits the broad application of their study results but the questions raised demand further research and consideration. 

“Let not the right side of your brain know what the left side doeth.”
~ George Bernard Shaw

By |October 16th, 2014|pathogenesis|0 Comments
Read More

It’s tough to make predictions, especially about the future but DWI reversal after tPA for stroke helps

Mark N Rubin, MD

Luby M, Warach SJ, Nadareishvili Z, and Merino JG. Immediate Changes in Stroke Lesion Volumes Post Thrombolysis Predict Clinical Outcome. Stroke. 2014


Mr. Berrawas right when he said “it’s tough to make predictions, especially about the future.” This is especially true in acute stroke practice, when all involved want to know exactly how things will go in the short, medium and long-term. However, in that individual patients frequently do not fit cleanly into our myriad risk scores because of the many, many moving parts in the cerebrovascular risk machine, the estimation power of combined clinical experience and risk stratification tools is imperfect.



The investigators of a recent study, as a substudy of a larger observational project, seek to augment our predictive powers with neuroimaging. This substudy was of patients presenting with acute stroke who were able to undergo acute multimodal MRI prior to intervention and then subsequent multimodal MRI scans at 2 and 24 hours post-intervention. These patients were then tracked for functional outcome out to at least three months, and the investigators went back to see which, if any, neuroimaging findings during that physiologically and clinically dynamic period were associated with a favorable outcome (e.g., modified Rankin score of 0-1, which basically means “functional independence”). Overall, in brief, binary multiple logistic regression analysis yielded younger age (per year decrease, OR=1.102, p=0.005, 95% CI: 1.03-1.18), admit NIHSS (per point decrease, OR=1.136, p=0.02, 95% CI: 1.020-1.264) and decrease in DWI volume at 24 hours (OR=1.126, p=0.008, 95% CI: 1.032-1.23) as significant predictors of favorable clinical outcome.

To complement the many studies that suggest DWI reversal with or without acute reperfusion therapy at some point in time during the early phase of acute stroke portends a good outcome for the patient, this investigation gives comprehensive, sequential, multimodal neuroimaging features that support a rational conclusion: if early reperfusion of ischemic tissue is achieved and sustained, less brain is injured and the patient does better.

Multimodal MRI is fascinating and provides valuable information to guide acute stroke therapy and prognostication, but is not widely available. Furthermore, only very few centers have the expertise to perform acute MRI before treatment without unnecessarily delaying systemic thrombolysis. All that being the case, this study should be read with an eye toward the principle of achieving and maintaining early reperfusion rather than implementation of multimodal MRI into pre-treatment acute stroke practice.

@MarkNRubinMD

By |October 15th, 2014|diagnosis and imaging|0 Comments
Read More

Stroke in Paradise: Implementing and Evaluating Prevention in a Afro-Caribbean Population in Martinique

Rizwan Kalani, MD

Olindo S, Chausson N, Mejdoubi M, Jeannin S, Rosillette K, Saint-Vil M, et al. Trends in Incidence and Early Outcomes in a Black Afro-Caribbean Population From 1999 to 2012: Etude Réalisée en Martinique et Centrée sur l’Incidence des Accidents vasculaires cérébraux II Study. Stroke. 2014

The Afro-Caribbean (AC) population of Martinique, a French Caribbean island, was found to have a high prevalence of vascular risk factors in a stroke epidemiology study conducted in 1998-1999 (ERMANCIA I). This led to initiation of community-based prevention strategies and a follow-up study using the same methodology (ERMANCIA II) was completed 13 years later (2011-2012). This program involved a large number of local healthcare professionals, both specialists and general practitioners, focusing on patient education regarding health behaviors (diet, physical activity) and engaging patients in tracking their risk factors (blood pressure, blood glucose). A dedicated stroke unit was also formed in 2003 at the local University Hospital of Fort de France.


In this article, Olindo et al evaluated this implementation program. In both ERMANCIA I and II, a comprehensive approach to determination of every stroke by a thorough evaluation of all hospital records and community resources was completed. Patient demographics, vascular risk factors, stroke subtypes, and outcomes were tracked.

The population of Martinique was stable between the study periods (380,000-390,000 people), and the majority is AC. First-ever stroke was identified in 544 AC patients during the ERMANCIA II study period, compared to 580 in 1998-1999. The age-standardised rates significantly decreased by 31% overall between the two evaluations (26% in males, 34% in females). Though the stroke rate of older people declined, a significant increase in those in a 35-44 year age group was noted, with an incidence rate ratio [IRR] 2.25 (95% confidence interval of 1.07-3.70). Frequencies of pre-morbid dyslipidemia, smoking, and atrial fibrillation were significantly higher in the 2011-2012 period; hypertension, diabetes, alcoholism, and coronary disease frequencies remained unchanged. When looking at subtypes, only ischemic stroke was found to be (statistically) significantly reduced over time, and only in women. Overall, the 30-day case-fatality ratio did not change, but the proportion of patients with a good outcome at 1 month (modified Rankin score ≤2) was significantly higher in ERMANCIA II (47% vs. 37.6% in 1998-1999, p=0.03). Along with decreasing stroke incidence, a stable high prevalence of hypertension and diabetes as well as increased proportion with dyslipidemia and tobacco use was found over the two time periods.

There were clearly limitations and challenges – the authors pointed out a lower proportion of patients evaluated by phone or from next of kin reports in ERMANCIA II compared to I. A noted decrease in the proportion of lacunar infarction could partly be explained by change in principal neuroimaging modality, from CT to more extensive MRI use in 2011-2012. This report describes stroke incidence in two periods of time; we cannot get an extensive understanding of the temporal trends. Nevertheless, there is a lot that can be learned from this work, which should be improved and emulated more extensively. It demonstrated that, even within this particular ethnic cohort, an increased stroke incidence was found in males compared to females aged 55-74 years. Just as is occurring elsewhere, the disturbing trend of increased smoking and obesity is contributing to higher stroke rates in younger population groups in Martinique.

This manuscript is another demonstration of the fact that efforts targeting prevention and appropriate management of stroke patients (patient education, dedicated stroke unit care) can reduce its incidence and improve outcomes. Further reduction of stroke incidence in Martinique will need to include addressing tobacco use and medication noncompliance in more effective ways. Much more can still be done. As shown in this report, large-scale epidemiology studies that track a condition, with its risk factors (and facilitators/barriers to addressing them), over time can be very powerful. If we were able and willing to do this for stroke in a coordinated manner at various levels (in a given city/state/country, regionally and even internationally), we could accelerate progress in addressing the enormous burden of cerebrovascular disease. Much can be learned from not only tracking health metrics (including risk factors, care quality, costs, and outcomes) over time, but also by comparing regions/populations where progress is being made and where it is not.

By |October 14th, 2014|epidemiology and genetics|0 Comments
Read More

T2’ Mapping draws a finer line between dead brain and penumbra

Daniel Korya, MD

Bauer S, Wagner M, Seiler A, Hattingen E, Deichmann R, Nöth U, and Singer OC. Quantitative T2′-Mapping in Acute Ischemic Stroke. Stroke. 2014

According to an article published in Nature Medicine in 2008 by Eng Lo, we are in the 4th decade of penumbral science. The first decade mostly focused on ischemic regulation of electrophysiology, cerebral blood flow and metabolism. The progress made in the second decade was largely due to molecular mechanisms that mediated penumbral cell death. The third decade is when neuroimaging became important clinically to isolate dead brain from salvageable penumbra. Now we are in the 4th decade, and indicated by Bauer and her colleagues, we may be closer to drawing a more precise boundary between dead brain and penumbra. 



As the authors point out, the aim of obtaining advanced imaging in the setting of acute stroke is to reliably detect the severity of ischemic changes with regard to irreversibly damaged and potentially salvageable brain tissue. That is why diffusion and perfusion weighted (DWI, PWI) imaging has become so important in recent clinical practice. The DWI provides a good estimate for the infarct core and the PWI is good at telling us the extent and severity of perfusion deficits. However, there is a small problem with PWI: although it tells you about perfusion to a certain area of the brain, it doesn’t tell you much about metabolic dysfunction.

We know from prior studies that the penumbra is an area around the stroke that has a cerebral blood flow of less than 20 ml/100g/min, but the oxygen consumption is preserved.  So, that means that the oxygen extraction fraction (OEF) must be significantly increased.  The T2’ relaxation time on the MRI is influenced by oxygen levels in hemoglobin since oxy-Hb is diamagnetic and deoxy-Hb is paramagnetic. So, it is ideal for telling us which parts of the ischemic brain are extracting more oxygen than others and therefor are still alive, but at risk. But, there is one problem with the T2’: it is highly susceptible to movement artifact and stroke patients (especially aphasic ones) are not very likely to follow commands and lay still in the MRI.

Bauer and her colleagues implemented a new “in-house developed” method to solve the problem of motion artifacts in T2’ imaging. They used this same method for subarachnoid hemorrhage patients and were successful, so they tried it in 11 ischemic stroke patients in this proof-of-concept study. 

The authors provided a technical, yet short, explanation of the method used to correct for motion. It’s a three-step algorithm: motion is detected with the correlation coefficient of the pixel-dependent exponential T2* fit. Then, T2* fitting is performed for three different input data sets based on resolution in the phase encoding direction. Finally, for each data set (100%, 50% and 25% resolutions), the best fit is chosen based on the constructed image with reduced movement. T2′ maps were calculated using the equation: 1/T2’ = 1/T2* – 1/T2. In order to exclude zero-voxels and voxels with cerebrospinal fluid, only T2` values between 1 ms and 300 ms were included for further analysis.

The bottom line is that it’s a complicated algorithm written in MATLAB that seemed to work before for SAH and it was worth a shot in ischemic stroke. 

The results were consistent with the concept of T2’ imaging and showed lower values within the ADC lesions (restricted diffusion) as well as lower values in the perfusion-restricted tissue (based on TTP). Accordingly, within the restricted diffusion lesions, the T2’ times were significantly slower than they were within the perfusion-restricted areas. 

This study is important to note for a few reasons: 1) It emphasizes the importance of MRI in acute ischemic stroke; 2) It suggests a solution to a known problem of motion artifact on T2’ imaging, and 3) It opens the door to potentially quantifying the level of ischemia in a stroke and could help in directing treatment. 

In an attempt to go “full circle” with this blog, it appears that Bauer and her colleagues have found a way to meld efforts made in the 4thdecade of penumbra science with those of the 1st decade: i.e. advanced neuroimaging can now detect oxygen consumption and metabolism in the ischemic brain more accurately.  
By |October 10th, 2014|diagnosis and imaging|1 Comment
Read More

Cracking ICD-9-CM Codes: Accuracy of Discharge Diagnoses in Stroke

Mark McAllister, MD


Jones SA, Gottesman RF, Shahar E, Wruck L, Rosamond WD. Validity of Hospital Discharge Diagnosis Codes for Stroke: The Atherosclerosis Risk in Communities Study. Stroke. 2014

Epidemiological estimates regarding stroke prevalence and mortality are often based on ICD-9-CM codes from hospital discharge. The accuracy of such statements is dependent on the codes actually corresponding to the labelled diagnosis. The authors sought to investigate the sensitivity and positive predictive values of ICD-9-CM codes for stroke and intracranial hemorrhage using diagnoses from the Atherosclerosis Risk in Communities (ARIC) study as the gold standard.


The ARIC study is comprised of nearly 16,000 patients in four communities in the US, and the database was searched for hospitalizations for ischemic strokes and intracranial hemorrhage. Strokes were identified by use of a stroke/hemorrhage related ICD-9-CM code, keywords in the discharge summary, or cerebral radiographic findings and validated by both computer algorithm and physician reviewer. Using this group of validated stroke and hemorrhage diagnoses the ICD-9-CM codes were compared.

Looking at AHA/ASA code groupings for ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage they found the positive predictive value to be 76% and sensitivity of 68%. An alternative grouping using more ICD codes increased sensitivity to 83%. These numbers are lower than previously published values and suggest stroke incidence may be underestimated based on ICD-9-CM codes.

By |October 9th, 2014|health care, outcomes, policy|0 Comments
Read More

Adding Puff to the Smoke: Combined Revascularization Surgery in Moyamoya Disease

Moyamoya disease is one that draws initial attention for neurologists due its intriguing name. “Puff of smoke” in Japanese, the disease is named after the angiographic appearance that patients with the condition have due to development of tiny collateral vessels to compensate for large artery stenosis usually occurring in bilateral ICAs, extending to the MCAs and ACAs. Revascularization procedures in the illness such as superficial temporal artery- middle cerebral artery (STA-MCA) bypass procedure and dural synangiosis have been used to improve collateral flow and appear to benefit patients in overall outcome. Usually patients are assigned to one type of bypass technique however the authors out of Korea attempted to combine the procedures of STA-MCA anastomosis and encephalodurogaleosynangiosis (EDGS) and see if these improved long term outcome in adult Moyamoya disease. 



The group performed STA-MCA bypass and EDGS on 77 hemisphere in 60 patients and followed the patients for short-term (approx. 6 months) and long-term (approx.. 5 year) periods and compared clinical status improvement based on Karnofsky Performance Scale (KPS) and modified Rankin Scale (mRS) measures. All procedures were performed by a single surgeon. They also quantitatively measured the revascularization area relative to the supratentorial area by using cerebral angiography as well as used SPECT scans with and without acetazolamide to be able to assess hemodynamic status for these patients both before and after the procedure. They found that clinical status by KPS and mRS improved from 81.1 (+/- 6.7) and 1.5 +/- 0.6 respectively to 96.2 (+/- 8.4) to 0.4 (+/- 0.7). Symptomatic clinical hyperperfusion syndrome was seen in 29.9% (23 operations, 21 patients).  The revascularization area improved from long-term follow up compared to short term (54.8% vs 44.2%, P<0.001). CBF also increased from preoperative values to the short term follow up period (P<0.001) and then became stable. Only 2 hemorrhagic events and 1 infarction were seen during follow up translating to symptomatic hemorrhage and infarction rates of 0.4%/person-year and 0.2%/person-year respectively. No patients died.

The combined revascularization technique makes sense as the STA-MCA bypass would provide immediate cerebral blood flow to the area of need and the EDGS would create more of a long-term benefit with slow forming collaterals over 3-6 months. Because no randomized control trials exist comparing surgical methods, and the inherent difficulties are present in comparing small retrospective trials given different surgical techniques, small number of patients, etc., the Korean group aimed to use a well-designed study to prove their point that this particular method can be used on adults with Moyamoya disease with relatively few complications and good outcomes and it appears they succeeded.  

@DrVikasNeuro

By |October 8th, 2014|treatment|0 Comments
Read More

Tetrahydrobiopterin supplementation in CADASIL patients

Duy Le, MD

De Maria R, Campolo J, Marina Frontali M, Taroni F, Federico A, Inzitari D, et al. Effects of Sapropterin on Endothelium-Dependent Vasodilation in Patients With CADASIL: A Randomized Controlled Trial. Stroke. 2014. 

De Maria et al bravely go where no study has gone before, and they should be commended for their efforts. They evaluated the effects of tetrahydrobiopterin (BH4), an essential co-factor for nitric oxide synthesis in endothelial cells, in CADASIL patients. This was a multi-center randomized, double blinded, placebo-controlled trial. Sixty-one CADASIL patients ages 30-65 were randomized to receive placebo or sapropterin, 200-400 mg BID, depending on their weight to target 5 mg/kg. Primary outcome was peripheral arterial tonometry (RH-PAT) performed at 24 months. The thought was that an increase RH-PAT value was determined to be a favorable outcome; meaning that there was more vasodilation and flow. The ITT population included 61 patients. RH-PAT was increased after 24 months in 37% of patients on sapropterin and 28% in placebo. However, when controlling for age, sex and clinical characteristics, improvement was not associated with the treatment arm.


Peripheral arterial tonometry was chosen by De Maria et al as an indirect and surrogate marker for amelioration of endothelial dysfunction seen in CADASIL based on the following thoughts 1) Previous studies in CADASIL patient describe impaired vasoreactivity in both cerebral and peripheral circulation. 2) PAT is independently shown to be associated with incident cardiovascular events in high risk patients. De Maria et al should be applauded for attempting to quantify a method which may be predictive of deterioration in CADASIL patients. However, while it does seem that there is a relationship in CADASIL patients between peripheral arteries and CNS arteries; we still do not know what the extent of that relationship is. More-over, how does that translate to development of disease progression or clinical deterioration?

Ultimately, this study should be lauded for its unique niche and massive undertaking to perform a RTC in such a rare disease entity. The primary outcome was not reached. If the thought is that there may be some improved benefit still of the sapropterin, this could be due either to a sample size that is too small (albeit very large for such a rare disease), a dosing that is too low, an effect is not actually seen in in the 24 month window, or the parameter used to measure improved outcome in these patients may actually be flawed. This was a phase-two study designed to show safety, and that outcome was achieved. While there is the potential to pursue a large scale RCT to evaluate for treatment effect, feasibility of such a process would be called into question as this is a rare disease.

By |October 7th, 2014|treatment|0 Comments
Read More

The continued search to better prognosticate patients receiving IV tPA External Validation of the BASIS and M1-BASIS in Thrombolysed Patients

Rajbeer Singh Sangha, MD

 Yeo LLL, Paliwal PR, Wakerley B, Khoo CM, Teoh HL, Ahmad A, et al. External Validation of the Boston Acute Stroke Imaging Scale and M1-BASIS inThrombolyzed Patients. Stroke. 2014

Current modalities of imaging and advances in technology have allowed physicians to visualize and locate arterial occlusion with excellent spatial resolution as well as view signs of early parenchymal insult.  The use of this information allows a better assessment of stroke severity and prognosis. The initial Boston Acute Stroke Imaging Scale (BASIS) is one such system that takes into account major arterial occlusion and parenchymal damage regardless of the clinical severity.  The authors of this study looked to validate BASIS and its sister scoring system, the M1-BASIS on pre thrombolysis scan, to determine their value in predicting the functional outcome in AIS patients treated with IV-tPA.



265 patients who underwent CTA of intra and extracranial vasculature before IV-tPA bolus were included in this study.  The results showed that the AUC for NIHSS alone was 0.728 to predict good outcomes while that of NIHSS+BASIS was 0.736.  Analysis showed that the 2 ROC curves were not significantly different.  The AUC for NIHSS + M1-BASIS was improved to 0.779, which was statistically significant (z= -0.1676, p = 0.047) when compared to the purely NIHSS ROC. This shows that M1-BASIS does have an additive effect with the NIHSS scale to prognosticate 3 month outcomes.

A scoring system for AIS must be simple, rapid, reproducible and possible in a variety of clinical settings. Detailed clinical rating instruments are commonly used in the settings of a clinical trial, and while these settings help in determining the efficacy of the instrument, it is often difficult to transition them to everyday practice.  As seen from the results of this study, M1-BASIS classification system correlates well with the clinical severity of stroke and can reliably prognosticate outcome in AIS patients treated with systemic thrombolysis.  Further analysis and evaluation should be performed to develop this score to be included in a paradigm which can be applied in a rapid fashion during AIS.  As technology and the ability of computers to exponentially combine information and perform calculations increases, the stroke community should continue to adapt and utilize these tools from the careful observations and analysis made from our studies.  

By |October 6th, 2014|diagnosis and imaging|0 Comments
Read More

Statistics bested by peer review in assessing operator performance in CREST trial

Chirantan Banerjee, MD

Howard G, Voeks JH, Meschia JF, Howard VJ, and Brott TJ. Picking the Good Apples: Statistics Versus Good Judgment in Choosing Stent Operators for a Multicenter Clinical Trial. Stroke. 2014

Despite the most compulsive adherence to pre, intra and post-procedure protocols, notable disparities in patient outcomes after surgery or procedures persist. Current guidelines specify that the procedural morbidity and mortality rate for procedure in asymptomatic carotid stenosis patients should be <3% and for symptomatic patients <6%.  The Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) was lauded for its low periprocedural stroke and death rates in asymptomatic and symptomatic carotid stenosis.
 


In this study, Howard et al. look at the data from this ever important trial and assess the scope of contribution by statistics in estimating an operator’s true complication rate. In CREST, a lead-in registry was used to assess potential stent operators, and average of 24 cases were evaluated for each operator by an interventional management committee. Thus, for an operator to have <3% complication rate, he could not have had even 1 complication among the 24 cases. Say an operator does 24 procedures without any complication, the conclusion that he will have a 0% complication rate may not be true, as the same operator may have 3 complications in his next 24 patients. So, the assessment of performance by this method is prone to error. In comparison, as the outcome is binary (complication v/s success) and thus has a binomial distribution, probability of exactly x complications out of the 24cases can be calculated. Thus, if we were to arbitrarily establish that an operator with say x complications out of 24 is a good operator, assessment can be made about the strength of this rule in identifying good and bad operators. The authors defined “error rate” as the sum of the expected percent of good operators (with a “true” 2% event rate) excluded, plus the percent of poor operators (with a true 6%, 8% or 10% event rate) included. They then go about calculating for each cohort of 100 operators with “true” complication rates ranging from 0 to 20%, who each perform 24 procedures, on average how many operators will have no complications, how many 1 complication, how many 2 complications, and so forth. This reveals that if statistics alone is used, assessing “good” and “bad” operators is fraught with significant error, unless the sample size is much higher. The fact that the committee was able to successfully include “good” operators strongly indicates the merit of combination of reviewing performance, volume and a subjective review of technique over pure statistics.   

When I performed a cursory search on how to quantify operator performance, I was surprised to learn that the data is very limited. An article in the NEJM last year found that technical skill of practicing bariatric surgeons in Michigan as assessed by peer rating of operative skill was associated with fewer postoperative complications.  At the end of the day, it probably relates to the fact that procedural skills are as much of an art as science. And only experience and peer evaluations may have the best odds at correctly triaging operator skill and technical prowess!

By |October 3rd, 2014|treatment|0 Comments
Read More