Framework for the Use of External Controls to Evaluate Treatment Outcomes in Precision Oncology Trials.

Advances in genomics have enabled anticancer therapies to be tailored to target specific genomic alterations. Single-arm trials (SATs), including those incorporated within umbrella, basket, and platform trials, are widely adopted when it is not feasible to conduct randomized controlled trials in rare biomarker-defined subpopulations. External controls (ECs), defined as control arm data derived outside the clinical trial, have gained renewed interest as a strategy to supplement evidence generated from SATs to allow comparative analysis. There are increasing examples demonstrating the application of EC in precision oncology trials. The prospective application of EC in conducting comparative studies is associated with distinct methodological challenges, the specific considerations for EC use in biomarker-defined subpopulations have not been adequately discussed, and a formal framework is yet to be established. In this review, we present a framework for conducting a prospective comparative analysis using EC. Key steps are (1) defining the purpose of using EC to address the study question, (2) determining if the external data are fit for purpose, (3) developing a transparent study protocol and a statistical analysis plan, and (iv) interpreting results and drawing conclusions on the basis of a prespecified hypothesis. We specify the considerations required for the biomarker-defined subpopulations, which include (1) specifying the comparator and biomarker status of the comparator group, (2) defining lines of treatment, (3) assessment of the biomarker testing panels used, and (4) assessment of cohort stratification in tumor-agnostic studies. We further discuss novel clinical trial designs and statistical techniques leveraging EC to propose future directions to advance evidence generation and facilitate drug development in precision oncology.

Effectiveness of a New Service Delivery Model for Management of Knee Osteoarthritis in Primary Care: A Cluster Randomized Controlled Trial.

OBJECTIVE: To evaluate the effectiveness and health costs of a new primary care service delivery model (the Optimising Primary Care Management of Knee Osteoarthritis [PARTNER] model) to improve health outcomes for patients with knee osteoarthritis (OA) compared to usual care. METHODS: This study was a 2-arm, cluster, superiority, randomized controlled trial with randomization at the general practice level, undertaken in Victoria and New South Wales, Australia. We aimed to recruit 44 practices and 572 patients age ≥45 years with knee pain for >3 months. Professional development opportunities on best practice OA care were provided to intervention group general practitioners (GPs). All recruited patients had an initial GP visit to confirm knee OA diagnosis. Control patients continued usual GP care, and intervention patients were referred to a centralized care support team (CST) for 12-months. Via telehealth, the CST provided OA education and an agreed OA action plan focused on muscle strengthening, physical activity, and weight management. Primary outcomes were patient self-reported change in knee pain (Numerical Rating Scale [range 0-10; higher score = worse]) and physical function (Knee Injury and Osteoarthritis Outcome Score activities of daily living subscale [range 0-100; higher score = better] at 12 months. Health care cost outcomes included costs of medical visits and prescription medications over the 12-month period. RESULTS: Recruitment targets were not reached. A total of 38 practices and 217 patients were recruited. The intervention improved pain by 0.8 of 10 points (95% confidence interval [95% CI] 0.2, 1.4) and function by 6.5 of 100 points (95% CI 2.3, 10.7), more than usual care at 12 months. Total costs of medical visits and prescriptions were $3,940 (Australian) for the intervention group versus $4,161 for usual care. This difference was not statistically significant. CONCLUSION: The PARTNER model improved knee pain and function more than usual GP care. The magnitude of improvement is unlikely to be clinically meaningful for pain but is uncertain for function.

Recursive Partitioning to Determine Order of Significance of Regional Metastasis Characteristics in Head and Neck Cutaneous Squamous Cell Carcinoma.

BACKGROUND: The order of significance of clinicopathologic characteristics for the prognosis of patients with regional metastases from head and neck cutaneous squamous cell carcinoma (HNcSCC) is not well characterized. This study aimed to understand the impact of the known characteristics, including the presence of immunosuppression, number of deposits, largest deposit size, location and laterality of deposits, and presence of extranodal extension (ENE) on overall survival (OS) and disease-specific survival (DSS). METHODS: A retrospective study of 366 patients treated with curative intent for HNcSCC with regional metastatic disease was undertaken using recursive partitioning analysis (RPA). RESULTS: Using RPA modeling, the study determined that number of metastatic deposits carried the highest impact for both OS and DSS, followed by largest deposit size. The presence of ENE and immunosuppression was less significant. CONCLUSIONS: The results from this study provide new evidence for identifying and stratifying high-risk patients with metastatic HNcSCC. This information will be valuable in determining future HNcSCC staging systems.

High-dose chemotherapy for relapsed germ cell tumours: outcomes in low-volume specialized centres.

OBJECTIVE: To report treatment patterns and survival outcomes of patients with relapsed and refractory metastatic germ cell tumours (GCTs) treated with high-dose chemotherapy (HDCT) and autologous stem-cell transplantation in low-volume specialized centres within the widely dispersed populations of Australia and New Zealand between 1999 and 2019. PATIENTS AND METHODS: We conducted a retrospective analysis of 111 patients across 13 institutions. Patients were identified from the Australasian Bone Marrow Transplant Recipient Registry. We reviewed treatment regimens, survival outcomes, deliverability and toxicities. Primary endpoints included overall (OS) and progression-free survival (PFS). Cox proportional hazards models were used to test the association of survival outcomes with patient and treatment factors. RESULTS: The median (range) age was 30 (14-68) years and GCT histology was non-seminomatous in 84% of patients. International Prognostic Factors Study Group (IPFSG) prognostic risk category was very low/low, intermediate, high and very high in 18%, 36%, 25% and 21% of patients, respectively. Salvage conventional-dose chemotherapy (CDCT) was administered prior to HDCT in 59% of patients. Regimens included paclitaxel, ifosfamide, carboplatin and etoposide (50%), carboplatin and etoposide (CE; 28%), carboplatin, etoposide and ifosfamide (CEI; 6%), carboplatin, etoposide and cyclophosphamide (CEC; 5%), CEC-paclitaxel (6%) and other (5%). With a median follow-up of 4.4 years, the 1-, 2- and 5-year PFS rates were 62%, 57% and 52%, respectively, and OS rates were 73%, 65% and 61%, respectively. There were five treatment-related deaths. Progression on treatment occurred in 17%. In a univariable analysis, worse International Germ Cell Cancer Collaborative Group (IGCCCG) and IPFSG prognostic groups were associated with inferior survival outcomes. An association of inferior survival was not found with the number of high-dose cycles received nor when HDCT was delivered after salvage CDCT. CONCLUSION: This large dual-national registry-based study reinforces the efficacy and deliverability of HDCT for relapsed and refractory metastatic GCT in low-volume specialized centres in Australia and New Zealand, with survival outcomes comparable to those found in international practice.

Distributional regression in clinical trials: treatment effects on parameters other than the mean.

BACKGROUND: The classical linear model is widely used in the analysis of clinical trials with continuous outcomes. However, required model assumptions are frequently not met, resulting in estimates of treatment effect that can be inefficient and biased. In addition, traditional models assess treatment effect only on the mean response, and not on other aspects of the response, such as the variance. Distributional regression modelling overcomes these limitations. The purpose of this paper is to demonstrate its usefulness for the analysis of clinical trials, and superior performance to that of traditional models. METHODS: Distributional regression models are demonstrated, and contrasted with normal linear models, on data from the LIPID randomized controlled trial, which compared the effects of pravastatin with placebo in patients with coronary heart disease. Systolic blood pressure (SBP) and the biomarker midregional pro-adrenomedullin (MR-proADM) were analysed. Treatment effect was estimated in models that used response distributions more appropriate than the normal (Box-Cox-t and Johnson's Su for MR-proADM and SBP, respectively), applied censoring below the detection limit of MR-proADM, estimated treatment effect on distributional parameters other than the mean, and included random effects for longitudinal observations. A simulation study was conducted to compare the performance of distributional regression models with normal linear regression, under conditions mimicking the LIPID study. The R package gamlss (Generalized Additive Models for Location, Scale and Shape), which implements maximum likelihood estimation for distributional regression modelling, was used throughout. RESULTS: In all cases the distributional regression models fit the data well, in contrast to poor fits obtained for traditional models; for MR-proADM a small but significant treatment effect on the mean was detected by the distributional regression model and not the normal model; and for SBP a beneficial treatment effect on the variance was demonstrated. In the simulation study distributional models strongly outperformed normal models when the response variable was non-normal and heterogeneous; and there was no disadvantage introduced by the use of distributional regression modelling when the response satisfied the normal linear model assumptions. CONCLUSIONS: Distributional regression models are a rich framework, largely untapped in the clinical trials world. We have demonstrated a sample of the capabilities of these models for the analysis of trials. If interest lies in accurate estimation of treatment effect on the mean, or other distributional features such as variance, the use of distributional regression modelling will yield superior estimates to traditional normal models, and is strongly recommended. TRIAL REGISTRATION: The LIPID trial was retrospectively registered on ANZCTR on 27/04/2016, registration number ACTRN12616000535471 .

Remodeling of the Feeding Arterial System After Surgery for Resection of Brain Arteriovenous Malformations: An Observational Study.

BACKGROUND: Avoiding the risk of postoperative hemorrhage after brain arteriovenous malformation (AVM) resection involves aggressive blood pressure control. Remodeling of the feeding arterial system is critical in reducing this risk. OBJECTIVE: To investigate factors predicting time to return to normal on digital subtraction angiography (DSA) after AVM resection. METHODS: For AVM in which the largest feeding artery (FA) on DSA was in the anterior circulation, the preoperative and postoperative diameter of the FA were compared with the diameter of the internal carotid artery (IC) immediately proximal to the posterior communicating artery. The preoperative FA/IC ratio (FA/IC preAVM) was compared with the first postoperative FA/IC ratio (FA/IC postAVM). Normal FA/IC ratio (FA/IC normal) was established from matched arteries in the contralateral hemisphere to the AVM. RESULTS: Eighty-six patients were analyzed for postoperative DSA performed a median 4 d after resection. From the interval-censored proportional hazards regression analysis, FA/IC preAVM (hazard ratio of 0.0006; 95% confidence interval: 0.00-0.21; P = .013) and maximum AVM diameter (hazard ratio of 0.47; 95% confidence interval: 0.23-0.95; P = .036) were significant in time to return to normal. These 2 factors were poorly correlated with each other (r = 0.41). AVMs with FA/IC preAVMs <0.57 combined with a diameter <3.0 cm normalize within 7 d in more than 50% of cases. Any other combination of ratio and size has fewer than 20% normalizing within 7 d (log rank P < .001). CONCLUSION: FA/IC preAVM and AVM size are both important in predicting the time taken for return to normal feeding arterial system on DSA after AVM resection.

Beyond mean modelling: Bias due to misspecification of dispersion in Poisson-inverse Gaussian regression.

In clinical trials one traditionally models the effect of treatment on the mean response. The underlying assumption is that treatment affects the response distribution through a mean location shift on a suitable scale, with other aspects of the distribution (shape/dispersion/variance) remaining the same. This work is motivated by a trial in Parkinson's disease patients in which one of the endpoints is the number of falls during a 10-week period. Inspection of the data reveals that the Poisson-inverse Gaussian (PiG) distribution is appropriate, and that the experimental treatment reduces not only the mean, but also the variability, substantially. The conventional analysis assumes a treatment effect on the mean, either adjusted or unadjusted for covariates, and a constant dispersion parameter. On our data, this analysis yields a non-significant treatment effect. However, if we model a treatment effect on both mean and dispersion parameters, both effects are highly significant. A simulation study shows that if a treatment effect exists on the dispersion and is ignored in the modelling, estimation of the treatment effect on the mean can be severely biased. We show further that if we use an orthogonal parametrization of the PiG distribution, estimates of the mean model are robust to misspecification of the dispersion model. We also discuss inferential aspects that are more difficult than anticipated in this setting. These findings have implications in the planning of statistical analyses for count data in clinical trials.

Comparing outcome scales for unruptured intracranial aneurysms: A prospective cohort study.

Outcomes on the modified Rankin Scale (mRS) are commonly used to guide and evaluate the management of unruptured intracranial aneurysms (uIA). However, the mRS is unlikely to measure all the relevant aspects of the clinical health of a patient. The current study therefore investigated the relationship between the mRS and additional measures of outcome. Between January 2011 and January 2016 patients with a new diagnosis of uIA were prospectively examined at referral and 12-month follow-up. Assessment included the Physical and Mental Component Scores of the Short Form 36 (SF-36), the computerized driver screening instrument DriveSafe (DS), and the mRS. Minimally Important Change (MIC) for each outcome measure was used to identify adverse outcomes for individual patients. A total of 128 patients (98 surgery; 30 untreated) completed the minimal dataset for analysis. In the surgical group, 6% (95% CI 3-14%) experienced morbidity at 12-months, as defined by the MIC for mRS. This risk rate increased to 51% (95% CI 41-61%) when defined as an MIC on any outcome. A combined MIC also identified a downgrade in outcomes, not detectable on the mRS, in 42% (95% CI 26-61%) of untreated patients. Correlation and regression analyses were unable to identify any significant relationships between the different outcomes instruments. In sum, there were considerably more adverse outcomes reported by quality of life (SF-36) and functional (DS) instruments than by the mRS for either treated or untreated uIA. To obtain a more complete representation of patient outcomes requires administration of a multi-dimensional assessment.

Measurement of Symptom Change Following Web-Based Psychotherapy: Statistical Characteristics and Analytical Methods for Measuring and Interpreting Change.

BACKGROUND: Accurate measurement of treatment-related change is a key part of psychotherapy research and the investigation of treatment efficacy. For this reason, the ability to measure change with accurate and valid methods is critical for psychotherapy. OBJECTIVE: The aims of this study were to (1) explore the underlying characteristics of depressive symptom change, measured with the nine-item Patient Health Questionnaire (PHQ-9), following psychotherapy, and (2) compare the suitability of different ways to measure and interpret symptom change. A treatment sample of Web-based psychotherapy participants (n=1098) and a waitlist sample (n=96) were used to (1) explore the statistical characteristics of depressive symptom change, and (2) compare the suitability of two common types of change functions: linear and proportional change. METHODS: These objectives were explored using hypotheses that tested (1) the relationship between baseline symptoms and the rate of change, (2) the shape of symptom score distribution following treatment, and (3) measurement error associated with linear and proportional measurement models. RESULTS: Findings demonstrated that (1) individuals with severe depressive baseline symptoms had greater reductions in symptom scores than individuals with mild baseline symptoms (11.4 vs 3.7); however, as a percentage measurement, change remained similar across individuals with mild, moderate, or severe baseline symptoms (50%-55%); (2) positive skewness was observed in PHQ-9 score distributions following treatment; and (3) models that measured symptom change as a proportional function resulted in greater model fit and reduced measurement error (<30%). CONCLUSIONS: This study suggests that symptom scales, sharing an implicit feature of score bounding, are associated with a proportional function of change. Selecting statistics that overlook this proportional change (eg, Cohen d) is problematic and leads to (1) artificially increased estimates of change with higher baseline symptoms, (2) increased measurement error, and (3) confounded estimates of treatment efficacy and clinical change. Implications, limitations, and idiosyncrasies from these results are discussed.

"Wish You Were Here": Examining Characteristics, Outcomes, and Statistical Solutions for Missing Cases in Web-Based Psychotherapeutic Trials.

BACKGROUND: Missing cases following treatment are common in Web-based psychotherapy trials. Without the ability to directly measure and evaluate the outcomes for missing cases, the ability to measure and evaluate the effects of treatment is challenging. Although common, little is known about the characteristics of Web-based psychotherapy participants who present as missing cases, their likely clinical outcomes, or the suitability of different statistical assumptions that can characterize missing cases. OBJECTIVE: Using a large sample of individuals who underwent Web-based psychotherapy for depressive symptoms (n=820), the aim of this study was to explore the characteristics of cases who present as missing cases at posttreatment (n=138), their likely treatment outcomes, and compare between statistical methods for replacing their missing data. METHODS: First, common participant and treatment features were tested through binary logistic regression models, evaluating the ability to predict missing cases. Second, the same variables were screened for their ability to increase or impede the rate symptom change that was observed following treatment. Third, using recontacted cases at 3-month follow-up to proximally represent missing cases outcomes following treatment, various simulated replacement scores were compared and evaluated against observed clinical follow-up scores. RESULTS: Missing cases were dominantly predicted by lower treatment adherence and increased symptoms at pretreatment. Statistical methods that ignored these characteristics can overlook an important clinical phenomenon and consequently produce inaccurate replacement outcomes, with symptoms estimates that can swing from -32% to 70% from the observed outcomes of recontacted cases. In contrast, longitudinal statistical methods that adjusted their estimates for missing cases outcomes by treatment adherence rates and baseline symptoms scores resulted in minimal measurement bias (<8%). CONCLUSIONS: Certain variables can characterize and predict missing cases likelihood and jointly predict lesser clinical improvement. Under such circumstances, individuals with potentially worst off treatment outcomes can become concealed, and failure to adjust for this can lead to substantial clinical measurement bias. Together, this preliminary research suggests that missing cases in Web-based psychotherapeutic interventions may not occur as random events and can be systematically predicted. Critically, at the same time, missing cases may experience outcomes that are distinct and important for a complete understanding of the treatment effect.

The role of embolization before surgery for Spetzler-Ponce Class B and C brain AVMs: a prospective cohort series.

BACKGROUND: There is uncertainty of the benefit of preoperative embolization for Spetzler-Ponce Class (SPC) B and C arteriovenous malformations of the brain (bAVM). We examined whether or not preoperative embolization reduces the risk of permanent neurological deficits in SPC B and C bAVM surgery. METHODS: A prospective bAVM database (between1989 and 2015) was analyzed by regression for factors associated with a new permanent neurological deficit arising as a consequence of surgery or preoperative embolization with a modified Rankin Scale (mRS) score >1 at 12 months after surgery (adverse outcome). RESULTS: From a cohort of 785 patients with bAVM, 277 patients with SPC B or C bAVM were planned for treatment by surgery with (N.=67) or without (N.=210) preoperative embolization. There were significant differences (embolization versus no embolization) in: permanent neurological deficits leading to a mRS>1 (45% versus 20%, P<0.01); permanent neurological deficits leading to a mRS>2 (22% versus 8.1%, P=0.04); perioperative transfusion of 2.5 liters of blood or more (31% versus 16%, P<0.01); and, delayed postoperative hemorrhage (19% versus 8.1%, P=0.01). Regression analysis identified the following factors to be associated with increased likelihood of an adverse outcome: infratentorial location (odds ratio 0.441, P=0.045); SPC C bAVM (OR=0.501, P=0.034); earlier rank order of surgery (OR=0.994, P<0.01); and, preoperative embolization (OR=0.313, P<0.01). CONCLUSIONS: The use of preoperative embolization does not reduce adverse outcomes in SPC B and C bAVM. The role of embolization in the preoperative management of complex bAVM by surgery deserves further study.

The Risk of Seizure Following Surgery for Brain Arteriovenous Malformation: A Prospective Cohort Study.

BACKGROUND: The evidence for the risk of seizures following surgery for brain arteriovenous malformations (bAVM) is limited. OBJECTIVE: To determine the risk of seizures after discharge from surgery for supratentorial bAVM. METHODS: A prospectively collected cohort database of 559 supratentorial bAVM patients (excluding patients where surgery was not performed with the primary intention of treating the bAVM) was analyzed. Cox proportional hazards regression models (Cox regression) were generated assessing risk factors, a Receiver Operator Characteristic curve was generated to identify a cut-point for size and Kaplan-Meier life table curves created to identify the cumulative freedom from postoperative seizure. RESULTS: Preoperative histories of more than 2 seizures and increasing maximum diameter (size, cm) of bAVM were found to be significantly (P < .01) associated with the development of postoperative seizures and remained significant in the Cox regression (size as continuous variable: P = .01; hazard ratio: 1.2; 95% confidence interval: 1.0-1.3; more than 2 seizures: P = .02; hazard ratio: 2.1; 95% confidence interval: 1.1-3.8). The cumulative risk of first seizure after discharge from hospital following resection surgery for all patients with bAVM was 5.8% and 18% at 12 mo and 7 yr, respectively. The 7-yr risk of developing postoperative seizures ranged from 11% for patients with bAVM ≤4 cm and with 0 to 2 preoperative seizures, to 59% for patients with bAVM >4 cm and with >2 preoperative. CONCLUSION: The risk of seizures after discharge from hospital following surgery for bAVM increases with the maximum diameter of the bAVM and a patient history of more than 2 preoperative seizures.

Deliberate employment of postoperative hypotension for brain arteriovenous malformation surgery and the incidence of delayed postoperative hemorrhage: a prospective cohort study.

OBJECTIVE The aim of this study was to examine the impact of deliberate employment of postoperative hypotension on delayed postoperative hemorrhage (DPH) for all Spetzler-Ponce Class (SPC) C brain arteriovenous malformations (bAVMs) and SPC B bAVMs ≥ 3.5 cm in diameter (SPC B 3.5+). METHODS A protocol of deliberate employment of postoperative hypotension was introduced in June 1997 for all SPC C and SPC B 3.5+ bAVMs. The aim was to achieve a maximum mean arterial blood pressure (BP) ≤ 70 mm Hg (with cerebral perfusion pressure > 50 mm Hg) for a minimum of 7 days after resection of bAVMs (BP protocol). The authors compared patients who experienced DPH (defined as brain hemorrhage into the resection bed that resulted in a new neurological deficit or that resulted in reoperation during the hospitalization for microsurgical bAVM resection) between 2 periods (prior to adopting the BP protocol and after introduction of the BP protocol) and 4 bAVM categories (SPC A, SPC B 3.5- [that is, SPC B < 3.5 cm maximum diameter], SPC B 3.5+, and SPC C). Patients excluded from treatment by the BP protocol were managed in the intensive care unit to avoid moderate hypertensive episodes. The pooled cases of all bAVM treated by surgery were analyzed to identify characteristics associated with the risk of DPH. These identified characteristics were then examined by multiple logistic regression analysis in both SPC B 3.5+ and SPC C cases. RESULTS From a cohort of 641 bAVMs treated by microsurgery, 32 patients with DPH were identified. Of those, 66% (95% CI 48-80) had a permanent new neurological deficit with a modified Rankin Scale score of 2-6. This included a mortality rate of 13% (95% CI 4.4-29). The BP protocol was used to treat 162 patients with either SPC B 3.5+ or SPC C. For SPC B 3.5+, there was no significant reduction in DPH with the introduction of the BP protocol (p = 0.77). For SPC C, there was a significant (p = 0.035) reduction of DPH from 29% (95% CI 13%-53%) to 8.2% (95% CI 3.2%-18%) associated with the introduction of the BP protocol. Multiple logistic regression analysis found that the absence of the BP protocol (p = 0.011, odds ratio 7.5, 95% CI 1.6-36) remained significant for the development of DPH in patients with SPC C bAVMs. CONCLUSIONS Treating patients with SPC C bAVMs with a protocol that lowers BP immediately after resection seems to reduce the risk of DPH. For SPC A and SPC B 3.5- bAVMs, there is unlikely to be a need to do more than avoid postoperative hypertension. For SPC B 3.5+ bAVMs, a larger number of patients would be required to test the absence of benefit of the BP protocol.

Microsurgery for Spetzler-Ponce Class A and B arteriovenous malformations utilizing an outcome score adopted from Gamma Knife radiosurgery: a prospective cohort study.

OBJECTIVE The purpose of this study was to adapt and apply the extended definition of favorable outcome established for Gamma Knife radiosurgery (GKRS) to surgery for brain arteriovenous malformations (bAVMs). The aim was to derive both an error around the point estimate and a model incorporating angioarchitectural features in order to facilitate comparison among different treatments. METHODS A prospective microsurgical cohort was analyzed. This cohort included patients undergoing embolization who did not proceed to microsurgery and patients denied surgery because of perceived risk of treatment. Data on bAVM residual and recurrence during long-term follow-up as well as complications of surgery and preoperative embolization were analyzed. Patients with Spetzler-Ponce Class C bAVMs were excluded because of extreme selection bias. First, patients with a favorable outcome were identified for both Class A and Class B lesions. Patients were considered to have a favorable outcome if they were free of bAVM recurrence or residual at last follow-up, with no complication of surgery or preoperative embolization, and a modified Rankin Scale score of more than 1 at 12 months after treatment. Patients who were denied surgery because of perceived risk, but would otherwise have been candidates for surgery, were included as not having a favorable outcome. Second, the authors analyzed favorable outcome from microsurgery by means of regression analysis, using as predictors characteristics previously identified to be associated with complications. Third, they created a prediction model of favorable outcome for microsurgery dependent upon angioarchitectural variables derived from the regression analysis. RESULTS From a cohort of 675 patients who were either treated or denied surgery because of perceived risk of surgery, 562 had Spetzler-Ponce Class A or B bAVMs and were included in the analysis. Logistic regression for favorable outcome found decreasing maximum diameter (continuous, OR 0.62, 95% CI 0.51-0.76), the absence of eloquent location (OR 0.23, 95% CI 0.12-0.43), and the absence of deep venous drainage (OR 0.19, 95% CI 0.10-0.36) to be significant predictors of favorable outcome. These variables are in agreement with previous analyses of microsurgery leading to complications, and the findings support the use of favorable outcome for microsurgery. The model developed for angioarchitectural features predicts a range of favorable outcome at 8 years following microsurgery for Class A bAVMs to be 88%-99%. The same model for Class B bAVMs predicts a range of favorable outcome of 62%-90%. CONCLUSIONS Favorable outcome, derived from GKRS, can be successfully used for microsurgical cohort series to assist in treatment recommendations. A favorable outcome can be achieved by microsurgery in at least 90% of cases at 8 years following microsurgery for patients with bAVMs smaller than 2.5 cm in maximum diameter and, in the absence of either deep venous drainage or eloquent location, patients with Spetzler-Ponce Class A bAVMs of all diameters. For patients with Class B bAVMs, this rate of favorable outcome can only be approached for lesions with a maximum diameter just above 6 cm or smaller and without deep venous drainage or eloquent location.

Complication-Effectiveness Analysis for Unruptured Intracranial Aneurysm Surgery: A Prospective Cohort Study.

BACKGROUND: The aim of intervention for unruptured intracranial aneurysms (UIAs) is safe, effective treatment. OBJECTIVE: To analyze a prospective database for variables influencing the risk of surgery to produce a risk model adjusting this risk for effectively treated aneurysms. METHODS: First, we identified variables to create a model from multiple logistic regression for complications of surgery leading to a 12-month modified Rankin Scale score >1. Second, we established the long-term cumulative incidence of freedom from retreatment or rupture (treated aneurysm) from Kaplan-Meier analysis. Third, we combined these analyses to establish a model of risk of surgery per effective treatment. RESULTS: One thousand twelve patients with 1440 UIA underwent 1080 craniotomies. We found that 10.1% (95% confidence interval [CI], 8.4-12.0) of craniotomies resulted in a complication leading to a modified Rankin Scale score >1 at 12 months. Logistic regression found age (odds ratio, 1.04; 95% CI, 1.02-1.06), size (odds ratio, 1.12; 95% CI, 1.09-1.15), and posterior circulation location (odds ratio, 2.95; 95% CI, 1.82-4.78) to be significant. Cumulative 10-year risk of retreatment or rupture was 3.0% (95% CI, 1.3-7.0). The complication-effectiveness model was derived by dividing the complication risk by the 10-year cumulative freedom from retreatment or rupture proportion. Risk per effective treatment ranged from 1% for a 5-mm anterior circulation UIA in a 20-year-old patient to 70% for a giant posterior circulation UIA in a 70-year-old patient. CONCLUSION: Complication-effectiveness analyses increase the information available with regard to outcome for the management of UIAs.

Factors Associated With Proximal Intracranial Aneurysms to Brain Arteriovenous Malformations: A Prospective Cohort Study.

BACKGROUND: The risk of hemorrhage from a brain arteriovenous malformation (bAVM) is increased when an associated proximal intracranial aneurysm (APIA) is present. Identifying factors that are associated with APIA may influence the prediction of hemorrhage in patients with bAVM. OBJECTIVE: To identify patient- and bAVM-specific factors associated with APIA. METHODS: We analyzed a prospective database of bAVMs for factors associated with the presence of APIA. Factors analyzed included age, sex, bAVM size, aneurysm size, circulation contributing to the bAVM, location of the aneurysm, deep venous drainage, and Spetzler-Ponce categories. Multiple logistic regression was performed to identify an association with APIA. RESULTS: Of 753 cases of bAVM with complete angiographic surveillance, 67 (9%) were found to have APIA. Older age (continuous variable; odds ratio, 1.04; 95% confidence interval, 1.02-1.05) and posterior circulation supply to the bAVM (odds ratio, 2.29; 95% confidence interval, 1.32-3.99) were factors associated with increased detection of APIA. The association of posterior circulation-supplied bAVM was not due to infratentorial bAVM location because 72% of posterior circulation APIAs were supplying supratentorial bAVM. CONCLUSION: APIAs appear to develop with time, as evident from the increased age for those with APIAs. Furthermore, they were more likely present in bAVMs supplied by the posterior circulation. This may be due to a difference in hemodynamic stress. ABBREVIATIONS: APIA, associated proximal intracranial aneurysmbAVM, brain arteriovenous malformationDSA, digital subtraction angiographySMG, Spetzler-Martin gradeSPC, Spetzler-Ponce category.

The Risk of Seizure After Surgery for Unruptured Intracranial Aneurysms: A Prospective Cohort Study.

BACKGROUND: We aimed to identify a group of patients with a low risk of seizure after surgery for unruptured intracranial aneurysms (UIA). OBJECTIVE: To determine the risk of seizure after discharge from surgery for UIA. METHODS: A consecutive prospectively collected cohort database was interrogated for all surgical UIA cases. There were 726 cases of UIA (excluding cases proximal to the superior cerebellar artery on the vertebrobasilar system) identified and analyzed. Cox proportional hazards regression models and Kaplan-Meier life table analyses were generated assessing risk factors. RESULTS: Preoperative seizure history and complication of aneurysm repair were the only risk factors found to be significant. The risk of first seizure after discharge from hospital following surgery for patients with neither preoperative seizure, treated middle cerebral artery aneurysm, nor postoperative complications (leading to a modified Rankin Scale score >1) was <0.1% and 1.1% at 12 months and 7 years, respectively. The risk for those with preoperative seizures was 17.3% and 66% at 12 months and 7 years, respectively. The risk for seizures with either complications (leading to a modified Rankin Scale score >1) from surgery or treated middle cerebral artery aneurysm was 1.4% and 6.8% at 12 months and 7 years, respectively. These differences in the 3 Kaplan-Meier curves were significant (log-rank P < .001). CONCLUSION: The risk of seizures after discharge from hospital following surgery for UIA is very low when there is no preexisting history of seizures. If this result can be supported by other series, guidelines that restrict returning to driving because of the risk of postoperative seizures should be reconsidered. ABBREVIATIONS: MCA, middle cerebral arterymRS, modified Rankin ScaleUIA, unruptured intracranial aneurysms.

Complication-Effectiveness Analysis for Brain Arteriovenous Malformation Surgery: A Prospective Cohort Study.

BACKGROUND: Intervention for brain arteriovenous malformations (bAVMs) should aim at treatment that is safe and effective. OBJECTIVE: To analyze a prospective database to derive the probability of neurological deficit and adjust this risk for effectively treated bAVMs (complication-effectiveness analysis [CEA]). METHODS: First, we calculated the percentage of surgical complications leading to a modified Rankin Scale >1 at 12 months after surgery for each Spetzler-Ponce class (SPC). Second, we performed a sensitivity analysis of these results by including bAVMs not undergoing surgery, to correct for bias. Third, we established the long-term cumulative incidence of freedom from recurrence from Kaplan-Meier analysis. Finally, we combined the results to calculate the risk of surgery per effective treatment in a complication-effectiveness analysis. RESULTS: Seven hundred seventy-nine patients underwent 641 microsurgical resections. Complications of surgery leading to a modified Rankin Scale >1 at 12 months occurred in 1.4% (95% confidence interval [CI]: 0.5-3.3), 20% (95% CI: 15-26), and 41% (95% CI: 30-52) of SPC A, SPC B, and SPC C, respectively. The cumulative 9-year freedom from recurrence was 97% for SPC A and 92% for other bAVMs. The 9-year CEA risk was 1.4% (credible range: 0.5%-3.4%) for SPC A, 22% to 24% (credible range: 16%-31%) for SPC B, and 45% to 63% (credible range: 33%-73%) for SPC C bAVM. CONCLUSION: CEA presents the treatment outcome in the context of efficacy and provides a basis for comparing outcomes from techniques with different times to elimination of the bAVM. ABBREVIATIONS: bAVM, brain arteriovenous malformationCEA, complication-effectiveness analysisCI, confidence intervalCTA, computerized tomographic angiographyDSA, digital subtraction angiographyMRA, magnetic resonance angiographymRS, modified Rankin ScaleSMG, Spetzler-Martin gradeSPC, Spetzler-Ponce class.

How to analyze the Visual Analogue Scale: Myths, truths and clinical relevance.

BACKGROUND AND AIMS: The Visual Analogue Scale (VAS) is a popular tool for the measurement of pain. A variety of statistical methods are employed for its analysis as an outcome measure, not all of them optimal or appropriate. An issue which has attracted much discussion in the literature is whether VAS is at a ratio or ordinal level of measurement. This decision has an influence on the appropriate method of analysis. The aim of this article is to provide an overview of current practice in the analysis of VAS scores, to propose a method of analysis which avoids the shortcomings of more traditional approaches, and to provide best practice recommendations for the analysis of VAS scores. METHODS: We report on the current usage of statistical methods, which fall broadly into two categories: those that assume a probability distribution for VAS, and those that do not. We give an overview of these methods, and propose continuous ordinal regression, an extension of current ordinal regression methodology, which is appropriate for VAS at an ordinal level of measurement. We demonstrate the analysis of a published data set using a variety of methods, and use simulation to compare the power of the various methods to detect treatment differences, in differing pain situations. RESULTS: We demonstrate that continuous ordinal regression provides the most powerful statistical analysis under a variety of conditions. CONCLUSIONS AND IMPLICATIONS: We recommend that in the situation in which no covariates besides treatment group are included in the analysis, distribution-free methods (Wilcoxon, Mann-Whitney) be used, as their power is indistinguishable from that of the proposed method. In the situation in which there are covariates which affect VAS, the proposed method is optimal. However, in this case, if the VAS scores are not concentrated around either extreme of the scale, normal-distribution methods (t-test, linear regression) are almost as powerful, and are recommended as a pragmatic choice. In the case of small sample size and VAS skewed to either extreme of the scale, the proposed method has vastly superior power to other methods.