Overtraining in neural networks that interpret clinical data

Backpropagation neural networks are a computer-based pattern-recognition method that has been applied to the interpretation of clinical data. Unlike rule-based pattern recognition, backpropagation networks learn by being repetitively trained with examples of the patterns to be differentiated. We describe and analyze the phenomenon of overtraining in backpropagation networks. Overtraining refers to the reduction in generalization ability that can occur as networks are trained. The clinical application we used was the differentiation of giant cell arteritis (GCA) from other forms of vasculitis (OTH) based on results for 807 patients (593 OTH, 214 GCA) and eight clinical predictor variables. The 807 cases were randomly assigned to either a training set with 404 cases or to a cross-validation set with the remaining 403 cases. The cross-validation set was used to monitor generalization during training. Results were obtained for eight networks, each derived from a different random assignment of the 807 cases. Training error monotonically decreased during training. In contrast, the cross-validation error usually reached a minimum early in training while the training error was still decreasing. Training beyond the minimum cross-validation error was associated with an increased cross-validation error. The shape of the cross-validation error curve and the point during training corresponding to the minimum cross-validation error varied with the composition of the data sets and the training conditions. The study indicates that training error is not a reliable indicator of a network's ability to generalize. To find the point during training when a network generalizes best, one must monitor cross-validation error separately.

Multicenter Trial of Stereotactic Body Radiation Therapy for Low- and Intermediate-Risk Prostate Cancer: Survival and Toxicity Endpoints

PURPOSE

The radiobiology of prostate cancer may favor the extreme hypofractionation inherent in stereotactic body radiation therapy (SBRT); however, data from a large multicenter study are lacking. We therefore examined the hypothesis that dose-escalated SBRT can be safely administered across multiple institutions, with favorable 5-year disease-free survival (DFS) rates compared with historical controls.

METHODS AND MATERIALS

Twenty-one centers enrolled 309 patients with prostate adenocarcinoma: 172 with low-risk (LR) and 137 with intermediate-risk (IR) disease. All were treated with a non-coplanar robotic SBRT platform using real-time tracking of implanted fiducials. The prostate was prescribed 40 Gy in 5 fractions of 8 Gy. We assessed toxicities using Common Terminology Criteria for Adverse Events (CTCAE) version 3 and biochemical failure using the "nadir + 2" definition. The study population yielded 90% power to identify excessive (>10%) rates of grade ≥3 genitourinary (GU) or gastrointestinal toxicities and, in the LR group, 80% power to show superiority in DFS over a 93% historical comparison rate.

RESULTS

At a median follow-up of 61 months, 2 LR patients (1.2%) and 2 IR patients (1.5%) experienced grade 3 GU toxicities, far below the 10% toxicity rate deemed excessive (upper limits of 95% confidence interval, 3.5% and 4.3%, respectively). No grade 4 or 5 toxicities occurred. All grade 3 toxicities were GU, occurring 11 to 51 months after treatment. For the entire group, the actuarial 5-year overall survival rate was 95.6% and the DFS rate was 97.1%. The 5-year DFS rate was 97.3% for LR patients (superior to the 93% DFS rate for historical controls; P = .0008; lower limit of 95% confidence interval, 94.6%) and 97.1% for IR patients.

CONCLUSIONS

Dose-escalated prostate SBRT was administered with minimal toxicity in this multi-institutional study. Relapse rates compared favorably with historical controls. SBRT is a suitable option for LR and IR prostate cancer.

Effects of continuous positive airway pressure on neurocognitive function in obstructive sleep apnea patients: The Apnea Positive Pressure Long-term Efficacy Study (APPLES)

STUDY OBJECTIVE

To determine the neurocognitive effects of continuous positive airway pressure (CPAP) therapy on patients with obstructive sleep apnea (OSA).

DESIGN, SETTING, AND PARTICIPANTS

The Apnea Positive Pressure Long-term Efficacy Study (APPLES) was a 6-month, randomized, double-blind, 2-arm, sham-controlled, multicenter trial conducted at 5 U.S. university, hospital, or private practices. Of 1,516 participants enrolled, 1,105 were randomized, and 1,098 participants diagnosed with OSA contributed to the analysis of the primary outcome measures.

INTERVENTION

Active or sham CPAP MEASUREMENTS: THREE NEUROCOGNITIVE VARIABLES, EACH REPRESENTING A NEUROCOGNITIVE DOMAIN: Pathfinder Number Test-Total Time (attention and psychomotor function [A/P]), Buschke Selective Reminding Test-Sum Recall (learning and memory [L/M]), and Sustained Working Memory Test-Overall Mid-Day Score (executive and frontal-lobe function [E/F])

RESULTS

The primary neurocognitive analyses showed a difference between groups for only the E/F variable at the 2 month CPAP visit, but no difference at the 6 month CPAP visit or for the A/P or L/M variables at either the 2 or 6 month visits. When stratified by measures of OSA severity (AHI or oxygen saturation parameters), the primary E/F variable and one secondary E/F neurocognitive variable revealed transient differences between study arms for those with the most severe OSA. Participants in the active CPAP group had a significantly greater ability to remain awake whether measured subjectively by the Epworth Sleepiness Scale or objectively by the maintenance of wakefulness test.

CONCLUSIONS

CPAP treatment improved both subjectively and objectively measured sleepiness, especially in individuals with severe OSA (AHI > 30). CPAP use resulted in mild, transient improvement in the most sensitive measures of executive and frontal-lobe function for those with severe disease, which suggests the existence of a complex OSA-neurocognitive relationship.

CLINICAL TRIAL INFORMATION

Registered at clinicaltrials.gov. Identifier: NCT00051363.

CITATION

Kushida CA; Nichols DA; Holmes TH; Quan SF; Walsh JK; Gottlieb DJ; Simon RD; Guilleminault C; White DP; Goodwin JL; Schweitzer PK; Leary EB; Hyde PR; Hirshkowitz M; Green S; McEvoy LK; Chan C; Gevins A; Kay GG; Bloch DA; Crabtree T; Demen WC. Effects of continuous positive airway pressure on neurocognitive function in obstructive sleep apnea patients: the Apnea Positive Pressure Long-term Efficacy Study (APPLES). SLEEP 2012;35(12):1593-1602.

Safety and efficacy of retreatment with a bioengineered hyaluronate for painful osteoarthritis of the knee: results of the open-label Extension Study of the FLEXX Trial

OBJECTIVE

To evaluate the safety of repeated intra-articular (IA) injections of Euflexxa® (1% sodium hyaluronate; IA--BioHA) for painful knee osteoarthritis (OA).

DESIGN

Participants who completed the randomized, double-blind, 26-week FLEXX Trial comparing IA-BioHA to IA saline (IA-SA) for knee OA(1) received three weekly IA-BioHA injections in a 26-week Extension Study. Adverse events (AEs) were recorded and the effect of treatment on knee pain was measured immediately following a 50-foot walk test using a 100 mm visual analog scale (VAS). Responder rate, Medical Outcomes Study Short Form 36 scores, Patient's Global Assessment, and intake of rescue medication were also evaluated.

RESULTS

The Extension Study included 433 subjects, 219 who received IA-BioHA and 214 who received IA-SA during the FLEXX Trial. Safety results from the Extension Study indicated that 43.4% (188/433) of subjects had AEs, of which 4.8% (21/433) were deemed treatment-related AEs. Two AEs in the Extension Study led to discontinuation, and no joint effusion was reported. Patients who continued with IA-BioHA in the Extension Study maintained their improvement from baseline, with an average reduction in pain in the VAS score of -3.5 mm. Patients initially treated with IA-SA in the FLEXX Trial also had a reduction in VAS score of -9.0 mm. Secondary efficacy variables also improved during the Extension Study.

CONCLUSIONS

Repeat injections of IA-BioHA were effective, safe, well tolerated, and not associated with an increase in AEs, such as synovial effusions. Additional symptom improvements were noted for subjects who received either IA-BioHA or IA-SA in the FLEXX Trial.

CLINICAL TRIAL REGISTRATION NUMBER

NCT00379236.

Plasma disc decompression compared with fluoroscopy-guided transforaminal epidural steroid injections for symptomatic contained lumbar disc herniation: a prospective, randomized, controlled trial

Object

Patients with radiculopathy, with or without back pain, often do not respond to conservative care and may be considered for epidural steroid injection therapy or a disc decompression procedure. Plasma disc decompression (PDD) using the Coblation SpineWand device is a percutaneous, minimally invasive interventional procedure. The purpose of this study was to evaluate clinical outcomes with PDD as compared with standard care using fluoroscopy-guided transforaminal epidural steroid injection (TFESI) over the course of 2 years.

Methods

This was a multicenter randomized controlled clinical study. Ninety patients (18–66 years old) who had sciatica (visual analog scale score ≥50) associated with a single-level lumbar contained disc herniation were enrolled. In all cases, their condition was refractory to initial conservative care and 1 epidural steroid injection had failed. Participants were randomly assigned to receive either PDD (46 patients) or TFESI (44 patients, up to 2 injections).

Results

The patients in the PDD Group had significantly greater reduction in leg pain scores and significantly improved Oswestry Disability Index and 36-Item Short Form Health Survey ([SF-36], physical function, bodily pain, social function, and physical components summary) scores than those in the TFESI Group. During the 2-year follow-up, 25 (56%) of the patients in the PDD Group and 11 (28%) of those in the TFESI Group remained free from having a secondary procedure following the study procedure (log-rank p = 0.02). A significantly higher percentage of patients in the PDD Group showed minimum clinically important change in scores for leg and back pain and SF-36 scores that exceeded literature-based minimum clinically important changes. Procedure-related adverse events, including injection site pain, increased leg or back pain, weakness, and lightheadedness, were observed in 5 patients in the PDD Group (7 events) and 7 in the TFESI Group (14 events).

Conclusions

In study patients who had radicular pain associated with a contained lumbar disc herniation, those patients treated with PDD had significantly reduced pain and better quality of life scores than those treated using repeated TFESI. In addition, significantly more PDD patients than TFESI patients avoided having to undergo a secondary procedure during the 2-year study follow-up.

Association between Genotype of the Serotonin Transporter-Linked Polymorphic Region of the Serotonin Transporter Gene and Age of Onset of Methamphetamine Use: a Preliminary Analysis

Early-onset methamphetamine use increases the lifetime prevalence of methamphetamine dependence. An earlier onset of methamphetamine use leads to greater damage to the terminal ends of serotonin neurons, more reduction in serotonin transporter (5-HTT) density, and an increased propensity toward further methamphetamine use. Because the 5-HTT-linked polymorphic region (5'-HTTLPR) within the promoter region of the 5-HTT gene leads to differential expression of the 5-HTT, we examined, for the first time, whether there is a differential association between the long (L) and short (S) alleles of the 5'-HTTLPR and the age of first methamphetamine use (AMU). The study included 120 methamphetamine-dependent adults of European descent. Diagnosis of methamphetamine dependence and AMU were collected using structured questionnaires, and the 5'-HTTLPR genotypes were determined using the polymerase chain reaction-restriction fragment length polymorphism method. Statistical analysis with the general linear model detected a significant interactive effect of 5'-HTTLPR genotypes (SS vs. L-carriers) and gender, associated with AMU (F = 3.99; p = 0.048). Further analysis of 5'-HTTLPR effects on AMU in males and females separately showed that the SS genotype compared with L-carriers had about two times greater risk of an earlier onset of methamphetamine use in men (hazard ratio = 1.839; 95% confidence interval = 1.042-3.246; p = 0.036) but not in women. Together, our findings in this preliminary study suggest a greater risk for earlier onset methamphetamine use associated with the SS genotype of the 5'-HTTLPR among methamphetamine-dependent Caucasian males.

A double-blind, randomized, saline-controlled study of the efficacy and safety of EUFLEXXA for treatment of painful osteoarthritis of the knee, with an open-label safety extension (the FLEXX trial)

OBJECTIVE

To report the FLEXX trial, the first well-controlled study assessing the safety and efficacy of Euflexxa (1% sodium hyaluronate; IA-BioHA) therapy for knee osteoarthritis (OA) at 26 weeks.

METHODS

This was a randomized, double-blind, multicenter, saline-controlled study. Subjects with chronic knee OA were randomized to 3 weekly intra-articular (IA) injections of either buffered saline (IA-SA) or IA-BioHA (20 mg/2 ml). The primary efficacy outcome was subject recorded difference in least-squares means between IA-BioHA and IA-SA in subjects' change from baseline to week 26 following a 50-foot walk test, measured via 100-mm visual analog scale (VAS). Secondary outcome measures included Osteoarthritis Research Society International responder index, Western Ontario McMaster University Osteoarthritis Index VA 3.1 subscales, patient global assessment, rescue medication, and health-related quality of life (HRQoL) by the SF-36. Safety was assessed by monitoring and reporting vital signs, physical examination of the target knee following injection, adverse events, and concomitant medications.

RESULTS

Five hundred eighty-eight subjects were randomized to either IA-BioHA (n = 293) or IA-SA (n = 295), with an 88% 26 week completion rate. No statistical differences were noted between the treatment groups at baseline. In the IA-BioHA group, mean VAS scores decreased by 25.7 mm, compared with 18.5 mm in the IA-SA group. This corresponded to a median reduction of 53% from baseline for IA-BioHA and a 38% reduction for IA-SA. The difference in least-squares means was -6.6 mm (P = 0.002). Secondary outcome measures were consistent with significant improvement in Osteoarthritis Research Society International responder index, HRQoL, and function. Both IA-SA and IA-BioHA injections were well tolerated, with a low incidence of adverse events that were equally distributed between groups. Injection-site reactions were reported by 1 (<1%) subject in the IA-SA group and 2 (1%) in the IA-BioHA group.

CONCLUSIONS

IA-BioHA therapy resulted in significant OA knee pain relief at 26 weeks compared with IA-SA. Subjects treated with IA-BioHA also experienced significant improvements in joint function, treatment satisfaction, and HRQoL.

A comparison of speculum and nonspeculum collection of cervicovaginal specimens for fetal fibronectin testing

OBJECTIVE

Current recommendations include the use of a vaginal speculum for fetal fibronectin specimen collection. This article evaluates the equivalency of nonspeculum methods for collecting fetal fibronectin specimens.

STUDY DESIGN

Two separate prospective studies of patients more than 22 weeks' gestation were performed at 2 institutions with similar hypotheses and methods. Two sequential specimens were collected on each patient: 1 with speculum and 1 with the nonspeculum method. The order of collection was reversed or randomized in both studies. Two alternative nonspeculum collection methods are described.

RESULTS

The 2 study sample sizes were 169 and 31. Comparison of the fetal fibronectin test results between the speculum and nonspeculum methods demonstrated greater than 95% agreement with an intraclass Kappa coefficient greater than 0.85 in both studies. The order of collection did not result in significantly different fetal fibronectin averages.

CONCLUSION

These studies demonstrate that there is excellent agreement between fetal fibronectin results obtained by speculum and nonspeculum collection methods.

Study of redox potential in cytochrome c covalently bound to terminal oxidase of alkaliphilic Bacillus pseudofirmus FTU

Spectroelectrochemistry was used to determine the midpoint redox potentials of heme cofactors of the caa3-type cytochrome oxidase from the alkaliphilic bacterium Bacillus pseudofirmus FTU. The apparent midpoint potentials (E(m)(app)) for the most prominent transitions of hemes a and a3 (+193 and +334 mV, respectively) were found to be similar to the values reported for other enzymes with high homology to the caa3-type oxidase. In contrast, the midpoint potential of the covalently bound cytochrome c (+89 mV) was 150-170 mV lower than in cytochromes c, either low molecular weight or covalently bound to the caa3 complex in all known aerobic neutralophilic and thermo-neutralophilic bacteria. Such an unusually low redox potential of the covalently bound cytochrome c of the caa3-type oxidase of alkaliphilic bacteria, together with high redox potentials of hemes a and a3, ensures more than twice higher difference in redox potentials inside the respiratory complex compared to the homologous mitochondrial enzyme. The energy released during this redox transition might be stored in the transmembrane H+ gradient even under low Deltap in the alkaline environment of the bacteria at the expense of a significant increase in DeltaG of the coupled redox reaction.

A preliminary randomized, double-blind, placebo-controlled study of the safety and efficacy of ondansetron in the treatment of methamphetamine dependence

Methamphetamine dependence is an increasing public health problem in the United States. No efficacious medication for methamphetamine dependence has been developed. As ondansetron, a 5-HT3 receptor antagonist and modulator of cortico-mesolimbic dopamine function, has been shown to reduce some of the rewarding effects of d-amphetamine in animal and human laboratory studies, we decided to test whether it would be superior to placebo at reducing methamphetamine use. In a preliminary, multi-site, randomized, double-blind, 8-wk controlled trial, 150 methamphetamine-dependent men and women received ondansetron (0.25 mg, 1 mg, or 4 mg b.i.d.) or placebo. Participants were assessed on several measures of methamphetamine use including urine methamphetamine level up to three times per week. As a psychosocial adjunct to the medication condition, cognitive behavioural therapy also was administered three times per week. Ondansetron was well tolerated and was less likely than placebo to be associated with serious adverse events. Nevertheless, none of the ondansetron doses was superior to placebo at decreasing any of the measures of methamphetamine use, withdrawal, craving, or clinical severity of methamphetamine dependence. Our preliminary results do not support the utility of ondansetron, at the doses tested, as a treatment for methamphetamine dependence. These findings should be viewed in light of the possibility that a less intensive cognitive behavioural therapy regimen might have yielded more positive results in this initial phase II trial exploring for the efficacy of ondansetron.

Effect of non-steroidal anti-inflammatory medications on the risk of amyotrophic lateral sclerosis

Inflammatory processes may be involved in the pathogenesis of amyotrophic lateral sclerosis (ALS). We examined the association of non-steroidal anti-inflammatory drugs (NSAIDs) with the risk of ALS in case-control study of incident cases (n = 111) conducted within the Kaiser Permanente Medical Care Program of Northern California during the years 1996-2000. Controls (n = 258) randomly selected from the same population were frequency matched by age and gender to the ALS cases. Information regarding use of NSAIDs (non-aspirin and aspirin) and three classes of 'control' medications was collected by in-person structured interview. Subjects who used medication at least twice a week for at least a month were classified as 'ever users'. Multivariable logistic regression models were adjusted for age, gender, history of osteoarthritis/rheumatoid arthritis and pain, and other medication use. Overall, there was no association between NSAID use and ALS; however, some sex differences were noted for non-aspirin NSAID use. Among men, non-aspirin NSAID use was associated with a two-fold increased risk of ALS (adjusted odds ratio (OR) 2.0, 95% confidence interval (CI) 1.0-3.9), whereas among women, non-aspirin NSAID use was not associated with increased ALS risk (adjusted OR 0.5, 95% CI 0.2-1.2). ALS risk was not associated with aspirin use or with 'control' medications. This study did not find any evidence to suggest that NSAID use reduces the risk of ALS. The observed sex differences with non-aspirin NSAID use could be due to chance or an unmeasured confounder.

Controlling for occasion-specific effects when assessing the test–retest reliability of self-report health questionnaires

OBJECTIVE

This study proposes a method for self-report health questionnaires to adjust test-retest reliability for changes during the test-retest interval based on an external measure, and to distinguish such changes from random response errors.

METHODS

In our application, eighty participants completed the Symptoms of Illness Checklist (SIC) on two occasions, two weeks apart, immediately before interviews given on each occasion by one of two physicians in a crossover design. The physician interview scores served as external measures, and structural equation modeling was used to estimate the parameters of a model that corrected for the occasion-specific effect of participants' responses using information from the interviews.

RESULTS

Correcting for changes in symptoms during the test-retest interval increased SIC test-retest reliability from .744 to .804 and significantly improved model fit (chi2(diff)(1) = 30.78, p < .001).

CONCLUSIONS

The results suggest methods that can improve the evaluation of self-report health questionnaire test-retest reliability by identifying changes using an external measure, and distinguishing these from random response errors; these increased the estimated SIC test-retest reliability and indicated that the SIC was indeed able to measure changes over the studied time interval. This method can be applied across a broad range of questionnaires.

Iliac fixation inhibits migration of both suprarenal and infrarenal aortic endografts

OBJECTIVE

To evaluate the role of iliac fixation in preventing migration of suprarenal and infrarenal aortic endografts.

METHODS

Quantitative image analysis was performed in 92 patients with infrarenal aortic aneurysms (76 men and 16 women) treated with suprarenal (n = 36) or infrarenal (n = 56) aortic endografts from 2000 to 2004. The longitudinal centerline distance from the superior mesenteric artery to the top of the stent graft was measured on preoperative, postimplantation, and 1-year three-dimensional computed tomographic scans, with movement more than 5 mm considered to be significant. Aortic diameters were measured perpendicular to the centerline axis. Proximal and distal fixation lengths were defined as the lengths of stent-graft apposition to the aortic neck and the common iliac arteries, respectively.

RESULTS

There were no significant differences in age, comorbidities, or preoperative aneurysm size (suprarenal, 6.0 cm; infrarenal, 5.7 cm) between the suprarenal and infrarenal groups. However, the suprarenal group had less favorable aortic necks with a shorter length (13 vs 25 mm; P < .0001), a larger diameter (27 vs 24 mm; P < .0001), and greater angulation (19 degrees vs 11 degrees ; P = .007) compared with the infrarenal group. The proximal aortic fixation length was greater in the suprarenal than in the infrarenal group (22 vs 16 mm; P < .0001), with the top of the device closer to the superior mesenteric artery (8 vs 21 mm; P < .0001) as a result of the 15-mm uncovered suprarenal stent. There was no difference in iliac fixation length between the suprarenal and infrarenal groups (26 vs 25 mm; P = .8). Longitudinal centerline stent graft movement at 1 year was similar in the suprarenal and infrarenal groups (4.3 +/- 4.4 mm vs 4.8 +/- 4.3 mm; P = .6). Patients with longitudinal centerline movement of more than 5 mm at 1 year or clinical evidence of migration at any time during the follow-up period comprised the respective migrator groups. Suprarenal migrators had a shorter iliac fixation length (17 vs 29 mm; P = .006) and a similar aortic fixation length (23 vs 22 mm; P > .999) compared with suprarenal nonmigrators. Infrarenal migrators had a shorter iliac fixation length (18 vs 30 mm; P < .0001) and a similar aortic fixation length (14 vs 17 mm; P = .1) compared with infrarenal nonmigrators. Nonmigrators had closer device proximity to the hypogastric arteries in both the suprarenal (7 vs 17 mm; P = .009) and infrarenal (8 vs 24 mm; P < .0001) groups. No migration occurred in either group in patients with good iliac fixation. Multivariate logistic regression analysis revealed that iliac fixation, as evidenced by iliac fixation length (P = .004) and the device to hypogastric artery distance (P = .002), was a significant independent predictor of migration, whereas suprarenal or infrarenal treatment was not a significant predictor of migration. During a clinical follow-up period of 45 +/- 22 months (range, 12-70 months), there have been no aneurysm ruptures, abdominal aortic aneurysm-related deaths, or surgical conversions in either group.

CONCLUSIONS

Distal iliac fixation is important in preventing migration of both suprarenal and infrarenal aortic endografts that have longitudinal columnar support. Secure iliac fixation minimizes the risk of migration despite suboptimal proximal aortic neck anatomy. Extension of both iliac limbs to cover the entire common iliac artery to the iliac bifurcation seems to prevent endograft migration.

Performance goals and endpoint assessments for clinical trials of femoropopliteal bare nitinol stents in patients with symptomatic peripheral arterial disease

OBJECTIVE

This analysis proposes safety and performance goals for prospective single-arm trials of bare nitinol stents to treat patients with debilitating claudication associated with femoropopliteal (FP) atherosclerotic lesions.

BACKGROUND

To date there have been no analyses of clinical trials data to set efficacy and safety benchmarks for new bare nitinol stents in the treatment of claudication from FP disease. Industry has been reluctant to sponsor studies of nitinol stents due to logistical barriers.

METHODS

VIVA Physician's, Inc. (VPI) analyzed subject-level data from the PTA control arm of three randomized FDA device trials conducted by industry. Subjects with Rutherford category 2-4 claudication and FP lesion lengths 4-15 cm with 12 month duplex ultrasound (DUS) assessment were identified. These data were combined with the results of a survey of the medical literature (1990-2006) for similar subjects.

RESULTS

Analysis of the industry derived control arm PTA data identified 116 patients (mean lesion length 8.7 cm) with a 12 month DUS defined FP patency of 28%. A similar cohort of 191 patients was identified from the medical literature in which the 12-month vessel patency equaled 37%; from these combined patient cohorts, expected vessel patency for PTA was estimated to equal 33%.

CONCLUSION

Based on the PTA performance efficacy rate of 33% derived from industry clinical trial data and the medical literature, and the requirement that the bare nitinol stent 12-month efficacy performance goal be set to equal twice this rate, the patency efficacy goal equals 66%. Additional information is provided on safety and other reporting standards and stent integrity evaluation for bare metal stents.

A combined superiority and non-inferiority approach to multiple endpoints in clinical trials

Treatment comparisons in clinical trials often involve multiple endpoints. By making use of bootstrap tests, we develop a new non-parametric approach to multiple-endpoint testing that can be used to demonstrate non-inferiority of a new treatment for all endpoints and superiority for some endpoint when it is compared to an active control. It is shown that this approach does not incur a large multiplicity cost in sample size to achieve reasonable power and that it can incorporate complex dependencies in the multivariate distributions of all outcome variables for the two treatments via bootstrap resampling.

Candidate early predictors for progression to joint damage in systemic juvenile idiopathic arthritis

OBJECTIVE

To assess if joint damage at 2 years after diagnosis in patients with systemic juvenile idiopathic arthritis (SJIA) can be predicted by clinical or laboratory features assessed up to 3 or 6 months after diagnosis.

METHODS

Medical records from 70 children were retrospectively reviewed. The primary outcome measure was presence of joint damage at 2 years after diagnosis (JD2) as defined by presence of erosions or fusion in one or more joints. Potential predictor variables for JD2 in the first 3 and 6 months after diagnosis consisted of the highest observed white blood cell count, platelet count, erythrocyte sedimentation rate, active joint count, and presence of symptomatic pulmonary or cardiac disease or macrophage activation syndrome, and treatment data.

RESULTS

The outcome of interest, JD2, was identified in 15/70 patients. Classification-tree analysis identified a pair of variables (highest observed platelet count and number of active joints) measured within the first 3 months after diagnosis that together predicted progression to JD2 with an estimated sensitivity of 87%, specificity of 82%, and positive predictive value of 57%. Multivariate logistic regression analyses at 3 months found that higher quantities of joints with active arthritis and early use of methotrexate (MTX) were factors significantly associated with increased odds of progression to JD2 (active joints odds ratio = 1.08, 95% CI 1.00-1.16, p = 0.04; MTX OR = 11.85, 95% CI 1.89-74.26, p = 0.01). Unsupervised cluster analysis identified 2 major phenotypes of patients at 3 months characterized by different ages at onset, acute phase markers, active joint counts, and presence of serositis. These phenotypes differed 3-fold in proportion of subjects progressing to JD2 (p < 0.05).

CONCLUSION

By 3 months after diagnosis, a clinical phenotype based on active joint count and platelet count may be prognostic of an increased risk of progression to JD2. Use of corticosteroids did not appear to change the risk of joint damage. In contrast, the presence of serositis appeared to be associated with decreased risk of joint damage.

Endovascular aneurysm repair at 5 years: does aneurysm diameter predict outcome?

OBJECTIVE

The appropriate size threshold for endovascular repair of small abdominal aortic aneurysms (AAA) is unclear. We studied the outcome of endovascular aneurysm repair (EVAR) as a function of preoperative aneurysm diameter to determine the relationship between aneurysm size and long-term outcome of endovascular repair.

METHODS

We reviewed the results of 923 patients treated in a prospective, multicenter clinical trial of EVAR. Small aneurysms were defined according to two size thresholds of 5.5 cm and 5.0 cm. Two-way analysis was used to compare patients with small aneurysms (<5.5 cm, n = 441) to patients with large aneurysms (> or =5.5 cm, n = 482). An ordered three-way analysis was used to compare patients with small AAA (<5.0 cm, n = 145), medium AAA (5.0 to 5.9 cm, n = 461), and large AAA (> or =6.0 cm, n = 317). The primary outcome measures of rupture, AAA-related death, surgical conversion, secondary intervention, and survival were compared using Kaplan-Meier estimates at 5 years.

RESULTS

Median aneurysm size was 5.5 cm. The two-way comparison showed that 5 years after EVAR, patients with small aneurysms (<5.5 cm) had a lower AAA-related death rate (1% vs 6%, P = .006), a higher survival rate (69% vs 57%, P = .0002), and a lower secondary intervention rate (25% vs 32%, P = .03) than patients with large aneurysms (> or =5.5 cm). Three-way analysis revealed that patients with small AAAs (<5.0 cm) were younger (P < .0001) and were more likely to have a family history of aneurysm (P < .05), prior coronary intervention (P = .003), and peripheral occlusive disease (P = .008) than patients with larger AAAs. Patients with smaller AAAs also had more favorable aortic neck anatomy (P < .004). Patients with large AAAs were older (P < .0001), had higher operative risk (P = .01), and were more likely to have chronic obstructive pulmonary disease (P = .005), obesity (P = .03), and congestive heart failure (P = .004). At 5 years, patients with small AAAs had better outcomes, with 100% freedom from rupture vs 97% for medium AAAs and 93% for large AAAs (P = .02), 99% freedom from AAA-related death vs 97% for medium AAAs and 92% for large AAAs (P = .02) and 98% freedom from conversion vs 92% for medium AAAs and 89% for large AAAs (P = .01). Survival was significantly improved in small (69%) and medium AAAs (68%) compared to large AAAs (51%, P < .0001). Multivariate Cox proportional hazards modeling revealed that aneurysm size was a significant independent predictor of rupture (P = .04; hazard ratio [HR], 2.195), AAA-related death (P = .03; HR, 2.007), surgical conversion (P = .007; HR, 1.827), and survival (P = .001; HR, 1.351). There were no significant differences in secondary intervention, endoleak, or migration rates between small, medium, and large AAAs.

CONCLUSIONS

Preoperative aneurysm size is an important determinant of long-term outcome following endovascular repair. Patients with small AAAs (<5.0 cm) are more favorable candidates for EVAR and have the best long-term outcomes, with 99% freedom from AAA death at 5 years. Patients with large AAAs (> or =6.0 cm) have shorter life expectancy and have a higher risk of rupture, surgical conversion, and aneurysm-related death following EVAR compared to patients with smaller aneurysms. Nonetheless, 92% of patients with large AAAs are protected from AAA-related death at 5 years. Patients with AAAs of intermediate size (5 to 6 cm) represent most of the patients treated with EVAR and have a 97% freedom from AAA-related death at 5 years.

A preliminary randomized, double-blind, placebo-controlled study of the safety and efficacy of ondansetron in the treatment of cocaine dependence

Prior studies have demonstrated inefficacy among dopamine receptor antagonists for treating cocaine dependence. An alternative approach would be to investigate the ability of indirect inhibitors of cortico-mesolimbic dopamine release, such as the 5-HT(3) receptor antagonist ondansetron, to reduce cocaine's reinforcing effects. We hypothesized that ondansetron might be more efficacious than placebo at reducing cocaine intake and promoting abstinence in cocaine-dependent individuals. In a pilot randomized, double-blind, 10-week controlled trial, 63 treatment-seeking, cocaine-dependent men and women received ondansetron (0.25 mg, 1.0 mg, or 4.0 mg twice daily) or placebo. Up to three times per week, participants were assessed on several measures of cocaine use, including urine benzoylecgonine. Cognitive behavioral therapy was administered weekly. Ondansetron was well tolerated, causing no serious adverse events. The ondansetron 4.0 mg group had the lowest dropout rate among all treatment groups and a greater rate of improvement in percentage of participants with a cocaine-free week compared with the placebo group (p = 0.02), whereas the ondansetron 1.0 mg group had a lower rate of improvement in percentage of weekly mean non-use days than did placebo recipients (p = 0.04). These results suggest the possibility of a non-linear dose-response function, with evidence supporting efficacy for the 4.0 mg group.

Effect of reproductive factors and postmenopausal hormone use on the risk of amyotrophic lateral sclerosis

OBJECTIVE

To examine the associations of reproductive factors and postmenopausal hormone use with the risk of amyotrophic lateral sclerosis (ALS) among women.

METHODS

This case-control study was conducted within the Kaiser Permanente Medical Care Program (KPMCP) of Northern California during the years 1996-2000. Among the 193 postmenopausal women, 62 were incident ALS cases and 131 were controls randomly selected from KPMCP members and frequency matched by age and respondent type (self versus proxy) to the cases. Statistical analyses were carried out using logistic regression.

RESULTS

Reproductive factors such as age at menarche, age at final menstrual period, parity, oral contraceptive use, and type of menopause (natural vs. hysterectomy with or without oophorectomy) were not associated with risk of ALS. Postmenopausal hormone use was positively, but not significantly, associated with the risk of ALS (adjusted OR 1.9, 95% CI 0.9-3.8).

CONCLUSIONS

Reproductive factors were not associated with ALS risk. There is no evidence that suggests a protective effect of postmenopausal hormone use against the development of ALS. However, due to insufficient power, we cannot rule out a possible increase in ALS risk associated with postmenopausal hormone use.

The Apnea Positive Pressure Long-term Efficacy Study (APPLES): rationale, design, methods, and procedures

STUDY OBJECTIVE

To assess the size, time course, and durability of the effects of long-term continuous positive airway pressure (CPAP) therapy on neurocognitive function, mood, sleepiness, and quality of life in patients with obstructive sleep apnea.

DESIGN

Randomized, double-blinded, 2-arm, sham-controlled, multicenter, long-term, intention-to-treat trial of CPAP therapy.

SETTING

Sleep clinics and laboratories at 5 university medical centers and community-based hospitals.

PATIENTS OR PARTICIPANTS

Target enrollment is 1100 randomly assigned subjects across 5 clinical centers.

INTERVENTIONS

Active versus sham (subtherapeutic) CPAP.

MEASUREMENTS AND RESULTS

A battery of conventional and novel tests designed to evaluate neurocognitive function, mood, sleepiness, and quality of life.

CONCLUSIONS

The Apnea Positive Pressure Long-term Efficacy Study (APPLES) is designed to study obstructive sleep apnea and test the effects of CPAP through a comprehensive, controlled, and long-term trial in a large sample of subjects with obstructive sleep apnea.

Effect of reproductive factors and postmenopausal hormone use on the risk of Parkinson disease

OBJECTIVE

Parkinson disease (PD) is less common in women possibly because of hormonal or reproductive influences. The objective of this study was to evaluate the associations of reproductive factors and postmenopausal hormone use with the risk of PD among postmenopausal women.

METHODS

Incident cases (n = 178) and randomly selected age-matched controls (n = 189) who were members of the Kaiser Permanente Medical Care Program (KPMCP) of Northern California participated in the study conducted during the years 1994 to 1995. Statistical analyses were carried out using logistic regression.

RESULTS

The association of postmenopausal hormone use with PD risk depended on the type of menopause. Among women with history of a hysterectomy with or without an oophorectomy, estrogen use alone was associated with a 2.6-fold increased risk (adjusted odds ratio (OR) 2.6, 95% CI: 1.1 to 6.1) and significant trends in the risk of PD were observed with increasing duration of estrogen use, but disease risk was not influenced by recency of use. In contrast, among women with natural menopause, no increased risk of PD was observed with hormone use (estrogen alone or a combined estrogen-progestin regimen). Early age at final menstrual period (44 years or younger) was associated with reduction in risk (adjusted OR 0.5, 95% CI: 0.3 to 1.0). Age at menarche and parity were not associated with the risk of PD.

CONCLUSION

Postmenopausal use of estrogen alone may increase the risk of Parkinson disease (PD) among women with a hysterectomy. Among women with natural menopause for whom the usual treatment is combined estrogen-progestin therapy, no increased risk of PD was observed.

A comprehensive assessment of the safety of intravenous methamphetamine administration during treatment with selegiline

Selegiline (L-deprenyl) is a selective irreversible monoamine oxidase B inhibitor shown to be effective in the treatment of Parkinson's and Alzheimer's diseases. Recent evidence suggests that selegiline may also be useful in treating specific aspects of cocaine and nicotine dependence, generating interest in this compound for the treatment of methamphetamine addiction. To investigate this, we performed a randomized, single-blind, placebo-controlled study to evaluate the safety of selegiline treatment (as compared to placebo), concurrent with intravenous methamphetamine (15 or 30 mg). Secondary study objectives included determinations of plasma levels of selegiline and its metabolites, evaluating whether selegiline administration altered the pharmacokinetics of methamphetamine or its metabolites, and evaluating whether selegiline treatment alters the subjective responses to methamphetamine. Twenty-four methamphetamine-dependent participants were randomized to treatment, and 9 of these (N = 5 selegiline, N = 4 placebo) completed the entire protocol. The principal finding from this study was that intravenous administration of moderate doses of methamphetamine was safely tolerated during treatment with selegiline. No participants had electrocardiogram changes, and there were no meaningful differences in any laboratory values either between groups at screening or as a result of the study procedures. In general, adverse events were mild or moderate, and no subjects were discontinued due to adverse events or serious adverse events. Selegiline treatment did not enhance any of the cardiovascular changes (heart rate, blood pressure) produced by methamphetamine administration. Selegiline treatment slightly increased methamphetamine associated "bad effects" but did not alter any other subjective effects. The elimination half-life of methamphetamine was approximately 12 h, and selegiline did not alter clearance of methamphetamine. The available data suggest that selegiline is likely to be safe if used as a pharmacotherapy for methamphetamine dependence.

Protective conditioning for acute graft-versus-host disease

BACKGROUND

Conditioning with total lymphoid irradiation plus antithymocyte serum protects mice against acute graft-versus-host disease (GVHD) after hematopoietic-cell transplantation. We tested this strategy in humans.

METHODS

Thirty-seven patients with lymphoid malignant diseases or acute leukemia underwent an experimental conditioning regimen with 10 doses of total lymphoid irradiation (80 cGy each) plus antithymocyte globulin, followed by an infusion of HLA-matched peripheral-blood mononuclear cells from related or unrelated donors who received granulocyte colony-stimulating factor.

RESULTS

Of the 37 transplant recipients, only 2 had acute GVHD after hematopoietic-cell transplantation. Potent antitumor effects in patients with lymphoid malignant diseases were shown by the change from partial to complete remission. In the transplant recipients who underwent conditioning with total lymphoid irradiation and antithymocyte globulin, the fraction of donor CD4+ T cells that produced interleukin-4 after in vitro stimulation increased by a factor of five, and the proliferative response to alloantigens in vitro was reduced, as compared with normal control subjects and control subjects who underwent conditioning with a single dose of total-body irradiation (200 cGy).

CONCLUSIONS

A regimen of total lymphoid irradiation plus antithymocyte globulin decreases the incidence of acute GVHD and allows graft antitumor activity in patients with lymphoid malignant diseases or acute leukemia treated with hematopoietic-cell transplantation.

Endovascular Repair or Surveillance of Patients with Small AAA

OBJECTIVE

To compare the outcome of patients with small abdominal aortic aneurysms (AAA) treated in a prospective trial of endovascular aneurysm repair (EVAR) to patients randomized to the surveillance arm of the UK Small Aneurysm Trial.

METHOD

All patients with small AAA (< or = 5.5 cm diameter) treated with a stent graft (EVARsmall) in the multicenter AneuRx clinical trial from 1997 to 1999 were reviewed with follow up through 2003. A subgroup of patients (EVARmatch) who met the age (60-76 years) and aneurysm size (4.0-5.5 cm diameter) inclusion criteria of the UK Small Aneurysm Trial were compared to the published results of the surveillance patient cohort (UKsurveil) of the UK Small Aneurysm Trial (NEJM 346:1445, 2002). Endpoints of comparison were aneurysm rupture, fatal aneurysm rupture, operative mortality, aneurysm related death and overall mortality. The total patient years of follow-up for EVAR patients was 1369 years and for UK patients was 3048 years. Statistical comparisons of EVARmatch and UKsurveil patients were made for rates per 100 patient years of follow up (/100 years) to adjust for differences in follow-up time.

RESULTS

The EVARsmall group of 478 patients comprised 40% of the total number of patients treated during the course of the AneuRx clinical trial. The EVARmatch group of 312 patients excluded 151 patients for age < 60 or > 76 years and 15 patients for AAA diameter < 4 cm. With the exception of age, there were no significant differences between EVARsmall and EVARmatch in pre-operative factors or post-operative outcomes. In comparison to the UKsurveil group of 527 patients, the EVARmatch group was slightly older (70 +/- 4 vs. 69 +/- 4 years, p = 0.009), had larger aneurysms (5.0 +/- 0.3 vs. 4.6 +/- 0.4 cm, p < 0.001), fewer women (7 vs. 18%, p < 0.001), and had a higher prevalence of diabetes and hypertension and a lower prevalence of smoking at baseline. Ruptures occurred in 1.6% of EVARmatch patients and 5.1% of UKsurveil patients; this difference was not significant when adjusted for the difference in length of follow up. Fatal aneurysm rupture rate, adjusted for follow up time, was four times higher in UKsurveil (0.8/100 patient years) than in EVARmatch (0.2/100 patient years, p < 0.001); this difference remained significant when adjusted for difference in gender mix. Elective operative mortality rate was significantly lower in EVARmatch (1.9%) than in UKsurveil (5.9%, p < 0.01). Aneurysm-related death rate was two times higher in UKsurveil (1.6/100 patient years) than in EVARmatch (0.8/100 patient years, p = 0.03). All-cause mortality rate was significantly higher in UKsurveil (8.3/100 patient years) than in EVARmatch (6.4/100 patient years, p = 0.02).

CONCLUSIONS

It appears that endovascular repair of small abdominal aortic aneurysms (4.0-5.5 cm) significantly reduces the risk of fatal aneurysm rupture and aneurysm-related death and improves overall patient survival compared to an ultrasound surveillance strategy with selective open surgical repair.

Cocaine Rapid Efficacy Screening Trial (CREST): a paradigm for the controlled evaluation of candidate medications for cocaine dependence

AIM

Development of effective medications for the treatment of cocaine dependence remains a major priority for the National Institute on Drug Abuse (NIDA) at the National Institutes of Health. The Cocaine Rapid Efficacy Screening Trial (CREST) paradigm was developed by the Division of Treatment Research and Development (DT R&D) at NIDA with the goal of enhancing pilot clinical trial validity when systematically assessing a range of medications and drug classes for potential utility in treatment of cocaine dependence.

DESIGN

CREST utilizes a randomized, controlled, parallel group, blinded methodology for comparing one or more marketed medications against a standard, pharmaceutical grade placebo. The trial design is comprised of a flexible 24-week screening/baseline period followed by randomization to an 8-week treatment period.

MEASURES

Standard measures of outcomes for the CREST included urinary benzoylecgonine (primary metabolite of cocaine), retention, cocaine craving, depression, clinical global impression and HIV-risk behaviors. In order to facilitate comparisons of data from the CREST studies across sites, drug classes and time, standardized procedures, measures and psychosocial counseling were used.

RESULTS

A total of 19 medications were evaluated in out-patient treatment research clinics in Boston, Cincinnati, Los Angeles, New York and Philadelphia.

CONCLUSIONS

Findings supported decisions to move forward three medications (cabergoline, reserpine, tiagabine) using full-scale, adequately powered, randomized placebo-controlled trial designs. Lessons learned from the CREST experience continue to shape cocaine pharmacotherapy trial design and execution.

Retrospective analyses of pooled data from CREST I and CREST II trials for treatment of cocaine dependence

AIM

To analyze pooled data from the Cocaine Rapid Evaluation Screening Trial (CREST). Pooling data from these small pilot trials into four major drug classes permitted data exploration for treatment and covariate effects with increased sample size.

DESIGN

Small pilot trials were conducted to screen fifteen medications as prospective treatments for cocaine dependence. Studies included a flexible 2-week to 4-week screening/baseline period followed by an 8-week randomized treatment condition. Participants were randomized equally to one of up to three active medications or placebo.

SETTING

Five Medications Development Research Units at the five academic centers of University of Cincinnati, New York University, University of Pennsylvania, University of California Los Angeles and Boston University.

PARTICIPANTS

The pooled data set consisted of 357 total subjects. Standardized inclusion and exclusion criteria were employed in subject selection to enhance consistency of cocaine-dependent study participants across all sites (see reports on individual trials in this supplement for details). All participants provided at least two urine samples that were positive for cocaine metabolite during a two-week period prior to being randomized.

INTERVENTION

All subjects in these trials, those randomized to placebo and active medications, received active treatment in the form of evidence-based cognitive behavioral therapy.

MEASURES

Quantitative urine benzoylecgonine (BE), self-report of cocaine use, and total Brief Substance Craving Scale (BSCS) scores were compared between each class of medication and its matched-placebo group.

FINDINGS

Regression analysis of pooled data did not identify any statistically significant differences between treatment and matched-placebo for any of the four classes. Exploration of the effects of baseline covariates indicated that gender and African American status were associated significantly with outcome. Female gender was consistently associated with poorer outcomes for medication and placebo groups, while the direction of association between African American status and outcome differed by treatment groups. Retention was also examined: dropout rates may have been somewhat higher for placebo than treatment groups during the early active-treatment period. Classification trees were used to identify characteristics of subjects who were abstinent for at least two weeks during the eight-week trial; only 4.0% of females while 17.9% of males achieved this criterion.

CONCLUSIONS

Results presented here may prove useful for planning future clinical trials for therapies targeting cocaine dependence.

Cocaine Rapid Efficacy Screening Trials (CREST): lessons learned

AIMS

The Cocaine Rapid Efficacy Screening Trials (CREST) were designed by the National Institute on Drug Abuse Division of Treatment Research and Development (NIDA, DT R&D) to rapidly screen a number of medications potentially useful for the treatment of cocaine dependence.

DESIGN

Each CREST trial was designed to compare several medications in a single trial against an unmatched placebo. The placebo group was included in each trial to avoid the nearly universal positive response to medications seen in open-label trials. In addition, a common set of procedures and outcome measures were employed throughout to increase comparability of results obtained from different trials and from different times.

PARTICIPANTS

In all, 18 medications were screened in seven different trials, conducted in four different sites throughout the United States involving 398 cocaine-dependent patients.

FINDINGS

Three medications were found to be promising enough to include in subsequent larger trials. Common statistical procedures for evaluating medications were developed to facilitate comparisons across sites and across time. A portion of the data were pooled and analyzed, which yielded some useful insights into cocaine dependence and its treatment. Finally, a review of individual trials together with the pooled analysis revealed several potential improvements for future screening trials.

CONCLUSIONS

Overall, the CREST trials proved to be useful for rapidly screening medications for treatment of cocaine dependence, but several modifications in design should be made before this framework is applied further.

A placebo-controlled screening trial of tiagabine, sertraline and donepezil as cocaine dependence treatments

AIMS

To conduct a preliminary evaluation of the safety and efficacy of tiagabine, sertraline or donepezil versus an unmatched placebo control as a treatment for cocaine dependence.

DESIGN

A 10-week out-patient study was conducted using the Cocaine Rapid Efficacy and Safety Trial (CREST) study design.

SETTING

This study was conducted at the Cincinnati Medication Development Research Unit (MDRU) and at an affiliated site in Dayton, Ohio.

PARTICIPANTS

Participants met Diagnostic and Statistical Manual version IV (DSM-IV) criteria for cocaine dependence. Sixty-seven participants were enrolled with 55 completing final study measures.

INTERVENTION

The targeted daily doses of medication were tiagabine 20 mg, sertraline 100 mg and donepezil 10 mg. All participants received 1 hour of manualized individual cognitive behavioral therapy on a weekly basis.

MEASUREMENTS

Primary outcome measures of efficacy included urine benzoylecgonine (BE) level, Cocaine Clinical Global Impression Scale-Observer and self-report of cocaine use. Safety measures included adverse events, ECGs, vital signs and laboratory tests.

FINDINGS

Subjective measures of cocaine dependence indicated significant improvement for all study groups. Generalized estimating equations analysis indicated that the tiagabine group showed a trend toward a significant decrease in urine BE level from baseline to weeks 5-8 (P = 0.10) and non-significant changes for the other study groups. No pattern of physical or laboratory abnormalities attributable to treatment with any of the medications was identified. There were three serious adverse events reported, none of which were related to study procedures.

CONCLUSIONS

The present findings suggest that tiagabine may be worthy of further study as a cocaine dependence treatment.

Efficacy screening trials of paroxetine, pentoxifylline, riluzole, pramipexole and venlafaxine in cocaine dependence

AIMS

The two studies presented here were conducted to assess the efficacy of paroxetine, pentoxifylline, riluzole, venlafaxine and pramipexole as medications for the treatment of cocaine dependence.

DESIGN

A multi-arm, modified blinded, placebo-controlled design was used.

SETTING

The studies were conducted at the Boston VA Healthcare System and the Boston University School of Medicine Medication Development Research Unit (MDRU).

PARTICIPANTS

Participants met criteria for cocaine dependence during a 2-week screening period.

INTERVENTION

Following random assignment to one of the treatment groups, subjects received active medication or placebo for 8 weeks in combination with cognitive behavioral counseling. In the first study the efficacy of the antidepressant paroxetine (20 mg daily), the phosphodiesterase inhibitor pentoxifylline (1200 mg daily) and the glutamate release inhibitor riluzole (100 mg daily) was assessed. The antidepressant venlafaxine (150 mg daily) and the dopamine agonist pramipexole (1.5 mg daily) were evaluated in the second study.

MEASUREMENTS

Urine benzoylecgonine (BE) concentrations, self-report of cocaine use and global impression scores served as primary outcome measures. Secondary measures included assessments of cocaine craving and psychiatric functioning. Adverse events were monitored during the treatment period.

FINDINGS

None of the active medications produced greater reductions in urine BE concentrations over the treatment period than did placebo. There were trends for BE levels to become reduced in the pentoxifylline group during the first 4 weeks of treatment and for Addiction Severity Index (ASI) drug composite scores to be lower in the pentoxyfylline group at end-point compared to the placebo group. Significant within-group reductions in reported cocaine use and craving were found for all treatment groups, but none of the active medications were superior to placebo on these measures. The accuracy of self-reported cocaine use declined over the study period. Overall, the active medications were well tolerated.

CONCLUSIONS

This study does not support the use of paroxetine, pentoxifylline, riluzole, venlafaxine or pramipexole for the treatment of cocaine dependence. However, these results need to be interpreted with caution because of the small size and lack of homogeneity of the experimental groups.

A medication screening trial evaluation of reserpine, gabapentin and lamotrigine pharmacotherapy of cocaine dependence

AIMS

To conduct a preliminary evaluation of the safety and efficacy of reserpine, gabapentin or lamotrigine versus an unmatched placebo control as a treatment for cocaine dependence.

DESIGN

A 10-week out-patient study using the Cocaine Rapid Efficacy and Safety Trial (CREST) study design.

SETTING

The study was conducted at the Cincinnati Medication Development Research Unit (MDRU).

PARTICIPANTS

Participants met Diagnostic and Statistical Manual version IV (DSM-IV) criteria for cocaine dependence. Sixty participants were enrolled, with 50 participants completing the final study measures.

INTERVENTION

The targeted daily doses of medication were reserpine 0.5 mg, gabapentin 1800 mg and lamotrigine 150 mg. All participants received 1 hour of manualized individual cognitive behavioral therapy on a weekly basis.

MEASUREMENTS

Primary outcome measures of efficacy included urine benzoylecgonine (BE) level, Cocaine Clinical Global Impression scale--observer and self-report of cocaine use. Safety measures included adverse events, electrocardiograms (ECGs), vital signs and laboratory tests.

FINDINGS

Subjective measures of cocaine dependence indicated significant improvement for all study groups. Urine BE results indicated a significant improvement for the reserpine group (P < 0.05) and non-significant changes for the other study groups. No pattern of physical or laboratory abnormalities attributable to treatment with any of the medications was identified. There were three serious adverse events reported, none of which were related to study procedures. The medications appeared to be tolerated well.

CONCLUSIONS

The present findings suggest that reserpine may be worthy of further study as a cocaine dependence treatment.

A nonparametric comparison of the effectiveness of treatments: a multivariate toxicity-penalized approach

In cancer clinical trials, the amount of treatment dose actually received by a patient may be limited by severe toxicity or lack of efficacy. A nonparametric approach is proposed for comparing the effectiveness of treatments based on the bivariate relationship defined by the doses at which efficacy and toxicity are observed to occur. Simulation studies are used to contrast the performance of the new procedure with the usual method of comparing percentages of patients who have effective results. Data from a randomized clinical trial of patients with metastatic nonseminomatous germ cell tumors are used to illustrate the method.

Energetics of alkalophilic representatives of the genus Bacillus

Cytochrome and lipid composition of membranes is considered as the attributes required for adaptation of the alkalophiles to alkaline conditions. Respiratory chains of alkalophilic representatives of the genus Bacillus are discussed. Special attention is paid to the features of the Na(+)-cycle of these bacteria and to the features determining halo- and alkalotolerant phenotype, which have been reported due to recent achievements in genomics.

Quantitative air-trapping analysis in children with mild cystic fibrosis lung disease

The purpose of this study was to compare quantitative computed tomography air trapping (AT) and pulmonary function measurements between subjects with mild cystic fibrosis lung disease (MCF; forced expiratory volume in 1 sec (FEV1) > 70% predicted) and normal age-matched controls. Quantitative AT measurements at different levels of expiration were evaluated. Ten subjects from the MCF group and 10 normal subjects underwent inspiratory and expiratory spirometer-triggered chest high-resolution computed tomography (HRCT) and pulmonary function tests. Six matched CT images were obtained at full inflation and at a lung volume near residual volume (nRV). Quantitative measurements of AT were determined by evaluating expiratory CT lung density and by the percent of segmented lung which demonstrated AT on expiratory scans. Percent AT was evaluated for all lung slices combined (global AT), and also by regional assessment. Additional comparisons of lung density and percent air trapping were made in 10 CF subjects with three matched axial HRCT images at lung volumes corresponding to full inflation, near functional residual capacity (nFRC), and nRV. All measurements of expiratory lung density in CF subjects were significantly lower and % AT significantly higher than normal controls. Significant correlations for all subjects were observed between % global AT and RV/TLC as well as forced expiratory flow between 25-75% of forced vital capacity (FEF(25-75)) % predicted. Pulmonary density measurements and % AT better discriminated differences between groups than PFTs. Measurements made on expiratory scans near FRC showed significantly higher values for AT than those made near RV.

Meniscal tear and extrusion are strongly associated with progression of symptomatic knee osteoarthritis as assessed by quantitative magnetic resonance imaging

BACKGROUND

The relation between knee meniscal structural damage and cartilage degradation is plausible but not yet clearly proven.

OBJECTIVES

To quantitate the cartilage volume changes in knee osteoarthritis using magnetic resonance imaging (MRI), and determine whether meniscal alteration predicts cartilage volume loss over time.

METHODS

32 patients meeting ACR criteria for symptomatic knee osteoarthritis were studied. MRI knee acquisitions were done every six months for two years. The cartilage volumes of different knee regions were measured. Three indices of structural change in the medial and lateral menisci were evaluated--degeneration, tear, and extrusion--using a semiquantitative scale.

RESULTS

24 patients (75%) had mild to moderate or severe meniscal damage (tear or extrusion) at baseline. A highly significant difference in global cartilage volume loss was observed between severe medial meniscal tear and absence of tear (mean (SD), -10.1 (2.1)% v -5.1 (2.4)%, p = 0.002). An even greater difference was found between the medial meniscal changes and medial compartment cartilage volume loss (-14.3 (3.0)% in the presence of severe tear v -6.3 (2.7)% in the absence of tear; p<0.0001). Similarly, a major difference was found between the presence of a medial meniscal extrusion and loss of medial compartment cartilage volume (-15.4 (4.1)% in the presence of extrusion v -4.5 (1.7)% with no extrusion; p<0.001).

CONCLUSIONS

Meniscal tear and extrusion appear to be associated with progression of symptomatic knee osteoarthritis.

Explant analysis of AneuRx stent grafts: relationship between structural findings and clinical outcome

OBJECTIVE

We reviewed the structural findings of explanted AneuRx stent grafts used to treat abdominal aortic aneurysms, and relate the findings to clinical outcome measures.

METHODS

We reviewed data for all bifurcated AneuRx stent grafts explanted at surgery or autopsy and returned to the manufacturer from the US clinical trial and worldwide experience of more than 33,000 implants from 1996 to 2003. Devices implanted for more than 1 month with structural analysis are included in this article. Explant results were analyzed in relation to cause of explantation and pre-explant evidence of endoleak, enlargement, or device migration.

RESULTS

One hundred twenty explanted stent grafts, including 37 from the US clinical trial, were analyzed. Mean implant duration was 22 +/- 13 months (range, 1-61 months). Structural abnormalities included stent fatigue fractures, fabric abrasion holes, and suture breaks. The mean number of nitinol stent strut fractures per explanted device was 3 +/- 4, which represents less than 0.2% of the total number of stent struts in each device. The mean number of fabric holes per explanted device was 2 +/- 3, with a median hole size of 0.5 mm(2). Suture breaks were seen in most explanted devices, but composed less than 1.5% of the total number of sutures per device. "For cause" explants (n = 104) had a 10-month longer implant duration (P =.007) compared with "incidental" explants (n = 16). "For cause" explants had more fractures (3 +/- 5; P =.005) and fabric holes (2 +/- 3; P =.008) per device compared with "incidental" explants, but these differences were not significant (P =.3) when adjusted for duration of device implantation. Among clinical trial explants the number of fabric holes in grafts in patients with endoleak (2 +/- 3 per device) was no different from those without endoleak (3 +/- 4 per device; P = NS). The number of fatigue fractures or fabric holes was no different in grafts in clinical trial patients with pre-explant aneurysm enlargement compared with those without enlargement. Pre-explant stent-graft migration was associated with a greater number of stent strut fractures (5 +/- 7 per device; P =.04) and fabric holes (3 +/- 3 per bifurcation; P =.03) compared with explants without migration. Serial imaging studies revealed inadequate proximal, distal, or junctional device fixation as the probable cause of rupture or need for conversion to open surgery in 86% of "for cause" explants. Structural device abnormalities were usually remote from fixation sites, and no causal relationship between device findings and clinical outcome could be established.

CONCLUSIONS

Nitinol stent fatigue fractures, fabric holes, and suture breaks found in explanted AneuRx stent grafts do not appear to be related to clinical outcome measures. Longer term studies are needed to confirm these observations.

Measurement of structural progression in osteoarthritis of the hip: the Barcelona consensus group

OBJECTIVE

To outline the best available method of measurement for detecting progression of osteoarthritis (OA) of the hip especially in therapeutic trials.

METHOD

A Medline search of articles related to progression of hip OA was performed. A group of experts met over a 1.5-day session to review available literature and new research. Specific questions were addressed in order to reach a consensus on measuring progression of OA of the hip.

RESULTS

Of the available surrogate measures, a single yearly standing or reclined antero-posterior plain radiograph of the pelvis with feet internally rotated 15-20 degrees, can be evaluated with the use of an atlas for joint space width (JSW, interbone distance). There should be a minimum JSW upon baseline screening that may be 1 or 2 mm. Digitization of films offers a slight reduction in variability of measurements. Progression of OA can be calculated by measurement of the JSW on paired and blinded films. A reduction of > or = 0.5 mm is greater than the 'minimum perceptible difference' as well as the variation of most imaging techniques, and represents a clinically relevant and significant reduction in the JSW. Narrowing of the superomedial or superolateral JSW may tend to progress more rapidly than other changes. In clinical trials, patients who discontinue the study treatment need to be followed after discontinuation, and an imputation strategy which provides unbiased estimates of both the treatment effect and its variance is an appropriate technique for intent-to-treat analysis.

CONCLUSION

For the development of new agents intended to prevent, retard, stabilize or reverse the progress of OA of the hip, the radiographic methodology presently available is adequate to detect changes in hip JSW of OA.

Quantitative magnetic resonance imaging evaluation of knee osteoarthritis progression over two years and correlation with clinical symptoms and radiologic changes

OBJECTIVE

To evaluate the change in osteoarthritic (OA) knee cartilage volume over a two-year period with the use of magnetic resonance imaging (MRI) and to correlate the MRI changes with radiologic changes.

METHODS

Thirty-two patients with symptomatic knee OA underwent MRI of the knee at baseline and at 6, 12, 18, and 24 months. Loss of cartilage volumes were computed and contrasted with changes in clinical variables for OA and with standardized semiflexed knee radiographs at baseline at 1 and 2 years.

RESULTS

Progression of cartilage loss at all followup points was statistically significant (P < 0.0001), with a mean +/- SD of 3.8 +/- 5.1% for global cartilage loss and 4.3 +/- 6.5% for medial compartment cartilage loss at 6 months, 3.6 +/- 5.1% and 4.2 +/- 7.5% at 12 months, and 6.1 +/- 7.2% and 7.6 +/- 8.6% at 24 months. Discriminant function analysis identified 2 groups of patients, those who progressed slowly (<2% of global cartilage loss; n = 21) and those who progressed rapidly (>15% of global cartilage loss; n = 11) over the 2 years of study. At baseline, there was a greater proportion of women (P = 0.001), a lower range of motion (P = 0.01), a greater circumference and higher level of pain (P = 0.05) and stiffness in the study knee, and a higher body mass index in the fast progressor group compared with the slow progressor group. No statistical correlation between loss of cartilage volume and radiographic changes was seen.

CONCLUSION

Quantitative MRI can measure the progression of knee OA precisely and can help to identify patients with rapidly progressing disease. These findings indicate that MRI could be helpful in assessing the effects of treatment with structure-modifying agents in OA.

Aneurysm enlargement following endovascular aneurysm repair: AneuRx clinical trial

OBJECTIVE

The purpose of this study was to determine the incidence and significance of aneurysm enlargement, with or without treatment, in relation to the primary end points of rupture, surgical conversion, aneurysm-related death, and survival following endovascular repair.

METHOD

Aneurysm (AAA) size changes and clinical outcome of all patients treated from 1997 through 1998 during the Phase II AneuRx multicenter clinical trial of endovascular AAA repair were reviewed. Aneurysm dimensions and the presence or absence of endoleak were determined by an independent core laboratory, with enlargement or shrinkage defined as a diameter change of 5 mm or more compared with baseline.

RESULTS

Among 383 patients (89% men, 11% women, age 73 +/- 9 years), with a mean device implant time of 36 +/- 11 months (median = 39 months), aneurysm diameter decreased from 5.7 +/- 1.0 at baseline to 5.2 +/- 1.0 at 3 years (P =.0001). A total of 46 patients (12%) experienced AAA enlargement, 199 patients (52%) had no change in AAA diameter, and 138 patients (36%) had a decrease in AAA diameter of 5 mm or more. Significant risk factors for enlargement included age (enlargement patients were 4 years older on average than patients with aneurysms that decreased in size; P =.002) and the presence of an endoleak (P <.001). Among patients with endoleak at any time, 17% had aneurysm enlargement, whereas only 2% of patients without endoleak had aneurysm enlargement (P <.001). Patients with enlargement were more likely to undergo secondary endovascular procedures and surgical conversions (P <.001). Twenty patients (43%) with enlargement underwent treatment, and 26 patients were untreated. There were two deaths following elective surgical conversion and one death in a patient with untreated enlargement and a type I endoleak. Three aneurysms ruptured: one with enlargement, one with no change, and one with a decrease in aneurysm size; all three aneurysms were larger than 6.5 cm. Kaplan-Meier analysis showed that freedom from rupture at 3 years was 98% with enlargement, 99% with no change, and 99% with decrease in AAA size (log-rank test, not significant). Freedom from AAA death at 3 years was 93% in patients with enlargement, 99% in no increase, and 99% in decrease (P =.005). Survival at 3 years was 86% with increase, 82% with no change, and 93% with decrease (P =.02).

CONCLUSIONS

Aneurysm enlargement following endovascular repair was not associated with an increased risk of aneurysm rupture or decrease in patient survival during a 3-year observation period. Aneurysm size rather than enlargement may be a more meaningful predictor of rupture. Close follow-up and a high re-intervention rate (43%) may account for the low risk of rupture in patients with enlargement. The long-term significance of aneurysm enlargement following endovascular repair remains to be determined.

Stent graft migration after endovascular aneurysm repair: importance of proximal fixation

OBJECTIVE

We reviewed the incidence of stent-graft migration after endovascular aneurysm repair in a prospective multicenter trial and identified factors that may predispose to such migration.

METHODS

All patients who received treatment during the course of the multicenter AneuRx clinical trial were reviewed for evidence of stent-graft migration over 5 years, from 1996 to 2001. Post-deployment distance from the renal arteries to the proximal end of the stent graft and the proximal fixation length (length of the infrarenal neck covered by the stent graft) were determined in patients for whom pre-procedure and post-procedure computed tomography scans were measured in an independent core laboratory.

RESULTS

Stent-graft migration was reported in 94 of 1119 patients, with mean time after device implantation of 30 +/- 11 months. Freedom from migration was 98.6% at 1 year, 93.4% at 2 years, and 81.2% at 3 years (Kaplan-Meier method). Subset (n = 387) analysis revealed that initial device deployment was lower in 47 patients with migration, as evidenced by a greater renal artery to stent-graft distance (1.1 +/- 0.7 cm), compared with 340 patients without migration (0.8 +/- 0.6 cm; P =.006) on post-implantation computed tomography scan. Proximal fixation length was shorter in patients with migration (1.6 +/- 1.4 cm) compared with patients without migration (2.3 +/- 1.4 cm; P =.005). There was significant variation in migration rate among clinical sites (P <.001), ranging from 0% to 30% (median, 8%), with a greater than twofold difference in migration rate between the lowest quartile (6%) and the highest quartile (15%) clinical sites. Univariate and multivariate analysis revealed that renal artery to stent-graft distance (P =.001) and proximal fixation length (P =.005) were significant predictors of migration, and that each millimeter increase in distance below the renal arteries increased risk for subsequent migration by 5.8% and each millimeter increase in proximal fixation length decreased risk for migration by 2.5%. Pre-implantation aortic neck length, neck diameter, degree of device oversizing, correct versus incorrect oversizing, device type (stiff vs flexible), placement of proximal extender cuffs at the original procedure, and post-procedure endoleak were not significant predictors of migration. Migration was treated with placement of extender modules in 23 patients and surgical conversion in 7 patients; 64 patients (68%) with migration have required no treatment.

CONCLUSIONS

Stent-graft migration among patients treated in the AneuRx clinical trial appears to be largely related to low initial deployment of the device, below the renal arteries, and short proximal fixation length. Significant variation in migration rate among clinical sites highlights the importance of the technical aspects of stent-graft deployment. Advances in intraoperative imaging and deployment techniques that have been made since completion of the clinical trial facilitate precision of device placement below the renal arteries and should increase proximal fixation length. Whether this, together with increased iliac fixation length, will result in lower risk for migration remains to be determined in long-term follow-up studies.

Composite spirometric-computed tomography outcome measure in early cystic fibrosis lung disease

With the advent of therapies aimed at young patients with cystic fibrosis, who have mildly reduced pulmonary function, the need for improved outcome measures that discriminate treatment effects has become important. Pulmonary function measurements or chest high-resolution computed tomography (HRCT) scores have been separately used to assess interventions. We evaluated these modalities separately and together during a treatment study to develop a more sensitive outcome measure. In a 1-year trial, 25 children randomized either to daily Pulmozyme or to normal saline aerosol were evaluated at randomization and at 3 and 12 months. Outcome variables were pulmonary function test (PFT) results, a global HRCT score, and a composite score incorporating PFTs and HRCT scoring. Regression analyses with generalized estimating equations permitted estimation of the difference in treatment effect between groups over time for each outcome. The largest difference in treatment effects observed at 12 months, measured by the percentage change from baseline, were with the composite total and maximal CT/PFT scores (35.4 and 30.4%), compared with mean forced expiratory flow during the middle half of the FVC (FEF25-75%) (13.0%) and total and maximal global HRCT scores (6.2%, 7.2%). The composite total and maximal CT/PFT scores were the most sensitive outcome measures for discriminating a treatment effect in children with cystic fibrosis with normal or mildly reduced pulmonary function during a 1-year trial of Pulmozyme.

Incidence of Parkinson's disease: variation by age, gender, and race/ethnicity

The goal of this study was to estimate the incidence of Parkinson's disease by age, gender, and ethnicity. Newly diagnosed Parkinson's disease cases in 1994-1995 were identified among members of the Kaiser Permanente Medical Care Program of Northern California, a large health maintenance organization. Each case met modified standardized criteria/Hughes diagnostic criteria as applied by a movement disorder specialist. Incidence rates per 100,000 person-years were calculated using the Kaiser Permanente membership information as the denominator and adjusted for age and/or gender using the direct method of standardization. A total of 588 newly diagnosed (incident) cases of Parkinson's disease were identified, which gave an overall annualized age- and gender-adjusted incidence rate of 13.4 per 100,000 (95% confidence interval (CI): 11.4, 15.5). The incidence rapidly increased over the age of 60 years, with only 4% of the cases being under the age of 50 years. The rate for men (19.0 per 100,000, 95% CI: 16.1, 21.8) was 91% higher than that for women (9.9 per 100,000, 95% CI: 7.6, 12.2). The age- and gender-adjusted rate per 100,000 was highest among Hispanics (16.6, 95% CI: 12.0, 21.3), followed by non-Hispanic Whites (13.6, 95% CI: 11.5, 15.7), Asians (11.3, 95% CI: 7.2, 15.3), and Blacks (10.2, 95% CI: 6.4, 14.0). These data suggest that the incidence of Parkinson's disease varies by race/ethnicity.

Comparison of the NovaSom QSG™, a new sleep apnea home-diagnostic system, and polysomnography

BACKGROUND

Obstructive sleep apnea (OSA) is a serious, common, and underdiagnosed disorder that challenges health care resources. While polysomnography (PSG) represents the standard diagnostic test for OSA, portable devices provide an alternative diagnostic tool when issues of cost, time, geographic availability, or other constraints pose impediments to in-lab testing. This study compares the NovaSom QSG, a new sleep apnea home diagnostic system, to PSG both in the laboratory and in the home.

METHODS

Fifty-one consecutive adults referred to the sleep lab for suspicion of OSA underwent one night of in-lab, simultaneous recording of PSG and NovaSom QSG in addition to using the NovaSom QSG at home for three nights. Two separate comparisons were made using the apnea-hypopnea index (AHI): in-lab PSG to in-lab NovaSom QSG and in-lab PSG to home NovaSom QSG.

RESULTS

Using a clinical cut-off of AHI=15, the sensitivity and specificity of the in-lab NovaSom QSG vs. PSG were 95% and 91%, respectively. For home NovaSom QSG vs. in-lab PSG, the sensitivity was 91% and specificity was 83%. The intra-class correlation coefficient for the agreement between three separate nights of NovaSom QSG home data was 0.88.

CONCLUSIONS

In a patient population suspected of having OSA, the NovaSom QSG demonstrated acceptable sensitivity and specificity both in the lab and self-administered in the home, when compared to PSG.

Reliability of a quantification imaging system using magnetic resonance images to measure cartilage thickness and volume in human normal and osteoarthritic knees

OBJECTIVE

The aim of this study was to evaluate the reliability of a software tool that assesses knee cartilage volumes using magnetic resonance (MR) images. The objectives were to assess measurement reliability by: (1) determining the differences between readings of the same image made by the same reader 2 weeks apart (test-retest reliability), (2) determining the differences between the readings of the same image made by different readers (between-reader agreement), and (3) determining the differences between the cartilage volume readings obtained from two MR images of the same knee image acquired a few hours apart (patient positioning reliability).

METHODS

Forty-eight MR examinations of the knee from normal subjects, patients with different stages of symptomatic knee osteoarthritis (OA), and a subset of duplicate images were independently and blindly quantified by three readers using the imaging system. The following cartilage areas were analyzed to compute volumes: global cartilage, medial and lateral compartments, and medial and lateral femoral condyles.

RESULTS

Between-reader agreement of measurements was excellent, as shown by intra-class correlation (ICC) coefficients ranging from 0.958 to 0.997 for global cartilage (P<0.0001), 0.974 to 0.998 for the compartments (P<0.0001), and 0.943 to 0.999 for the condyles(P<0.0001). Test-retest reliability of within-reader data was also excellent, with Pearson correlation coefficients ranging from 0.978 to 0.999 (P<0.0001). Patient positioning reliability was also excellent, with Pearson correlation coefficients ranging from 0.978 to 0.999 (P<0.0001).

CONCLUSIONS

The results of this study establish the reliability of this MR imaging system. Test-retest reliability, between-reader agreement, and patient positioning reliability were all extremely high. This study represents a first step in the overall validation of an imaging system designed to follow progression of human knee OA.