Siddiqui and Zivadinov, Supplement Page 1

Siddiqui and Zivadinov, Supplement Page 1

Supplementary Appendices

Supplement to Siddiqui AH, Zivadinov R, Benedict RHB, et al. Prospective Randomized Trial of Venous Angioplasty in MS (PREMiSe)

These appendices have been provided by the authors to give readers additional information about their work.

Appendix e-1

Noninvasive screening

After signing an informed consent, patients needed to fulfill noninvasive duplex extracranial and transcranial venous hemodynamic (VH) criteria,1 to qualify for the invasive part of the study. Duplex examination was performed on a color-coded duplex scanner (MyLab 25; Esaote-Biosound, Irvine, California) equipped with a 2.5- to 10-Mhz transducer to examine internal jugular vein (IJV) morphology and hemodynamics. A subject was considered positive at screening in phase 1 if ≥2 chronic cerebrospinal venous insufficiency (CCSVI) VH criteria were fulfilled; and in phase 2, if ≥2 CCSVI extracranial VH criteria (1, 3, 4, and 5) were fulfilled. Duplex examination was performed by two trained technologists (KM and VV) with documented reproducibility and blinded to patient characteristics.1-3 An independent neuroradiologist (KD) read all duplex study examinations in a blinded manner.

Appendix e-2

Diagnostic catheter venography

The details of diagnostic catheter venography (CV) in the PREMiSe study are reported elsewhere.4Under conscious sedation with local anesthesia, an 8-French sheath was inserted using a modified Seldinger technique into the common femoral vein. Through this sheath, a guide catheter (5-French, 90-cm-long Head Hunter, Terumo Europe, Leuven, Belgium) was advanced through the inferior vena cava, across the right atrium into the superior vena cava. Catheterization proceeded to the azygous vein outlet into the superior vena cava. With the help of a hydrophilic guidewire (0.035-inch diameter Radiofocus Guide Wire M, Terumo Europe), the catheter was advanced inside the azygous vein until it neared the confluence with the hemi-azygous vein. An autoinjector was used to instill 9 ml of contrast medium (Visipaque, iodixanol, GE HealthCare; 270 mg/mL) at a constant rate of 3 ml/sec. Subtraction digital CV of the azygous vein was completed, with a right posterior oblique projection (range, 15°-25°) and an extended recording time. In this way, it was possible to achieve complete opacification of the system of origin of the azygous vein and hemi-azygous veins up to the ascending lumbar veins. Subsequently, the right internal jugular vein (IJV) and left IJV were approached, in that order. With the help of the guidewire, the catheter was moved inside each IJV up to the junction with the jugular bulb (skull base). Contrast medium (12ml) was injected at a constant rate of 3 ml/sec. Subtraction digital CV of each IJV was completed with an anterior-posterior projection and an extended recording time.

The ≥50% reduction in azygous vein or IJV lumen diameter detected on CV was also confirmed by use of intravascular ultrasound (IVUS) (Eagle Eye platinum catheter -20 MHz probe;Volcano s5/s5i Imaging system; Volcano, San Diego, CA), as previously reported.4 IVUS was not consistently performed across suspected areas of stenosis (≥50% narrowing) in phase 1 patients; however, it was consistently performed in all phase 2 patients, independently of the narrowing level on CV, to identify venous abnormalities. All IVUS images were read by two interpreters (AHS and YK). Stenosis documented by IVUS was calculated as the ratio between the minimal diameter of the vein in any of the axial images and the maximal diameter of the vein in any of the images. All identified stenoses were confirmed to be structural and not physiological by asking the patient to perform a Valsalva maneuver during the IVUS study.

Appendix e-3

Procedural RANDOMIZATION AND Blinding

Randomization in phase 2 was performed by an independent statistician in 1:1 fashion using sealed and numbered envelopes with predetermined treatments (10 angioplasty, 10 sham angioplasty). The envelopeswere randomly picked after the angiogram was completed and demonstrated at least one significant stenosis. Therefore, at the end of the study, 20 patients were randomized to the two arms. No preplanned replacement for subjects not fulfilling invasive screening criteria was included in the protocol.

No staff member from the safety, imaging, or clinical evaluation arms of this study was present during the treatment procedure. The operating room staff received training about the blinding requirements and avoided any loud procedure-related conversation. Relatively loud music per patient’s choice was played to further distract sedated patients so that procedural conversation was inaudible. X-ray shields were covered with opaque sterile covers and monitors were angled away from patients to prevent them from observing images of their procedure. All patients received a rigorous sternal rub (painful stimulus) upon insertion of the angioplasty balloon, regardless of inflation. A blindness assessment survey was administered the following morning prior to discharge. The survey results showed that 90% of phase 2 patients confirmed that they did not know whether they received angioplasty or the sham endovascular procedure.

Appendix e-4: Endovascular Procedures

Venous angioplasty

Patients were heparinized to confirm an activated clotting time of at least 250 seconds (s). A noncompliant balloon with nominal diameter of at least 80% of the proximal vein (of interest) was placed across the stenosis over a 0.035-inch glide wire. Inflation proceeded slowly at a rate of 1 atmosphere per 30s until nominal pressure was reached (8-12 atmospheres). The dilated balloon was left in place for 5 minutes and then deflated at a rate of 1 atmosphere per 15s. Once the balloon was completely deflated, it was withdrawn and the diagnostic catheter reintroduced over an exchange wire to perform a postprocedure selective venogram and assess residual stenosis. The goal of the angioplasty was to restore the venous outflow stricture to 50% of normal proximal venous diameter. Additional angioplasty was performed, if >50% residual stenosis remained. Once adequate angioplasty had been performed, the catheter, wire, and sheath were removed and the venous access site at the level of the common femoral vein was compressed using manual compression for 20 minutes followed by placement of a nonocclusive vascular clamp for 1 hour.

Postprocedurally, patients were admitted to a monitored (constant apnea and cardiac monitors) unit for observation of any immediate adverse events (AEs). The groin was inspected every 4 hours until discharge. At 4 hours postprocedure, if there was no evidence of a growing groin hematoma, the patients received the first dose of enoxaparin sodium, 30mg subcutaneously, which was continued on a once-daily basis for 3 weeks (drug provided at discharge). In addition, a daily dose of aspirin, 81mg, was given starting on the day of the procedure for a total of 3 weeks. Patients were observed overnight and discharged the next morning if there were no AEs.

Sham procedure

As above, except the balloon was inserted but not inflated in the sham-procedure group.

Appendix e-5: Outcomes

Safety outcomes

Given the complexity and duration (approximately 90min) of the invasive diagnostic and angioplasty procedures and need for additional experience, phase 1 was planned to: a) gain experience with use of catheter venography (CV) for accurate invasive diagnostic assessment of extracranial venous drainage, b) determine preliminary safety of the extracranial venous angioplasty procedure and related adverse events (AEs), and c) determine safety of the angioplasty procedure relative to magnetic resonance imaging (MRI) outcomes (presence of ≥5 new contrast-enhancing lesions on follow-up MRI in 5 consecutive patients was established as a safety rule). Upon completion of phase 1 with an acceptable AE rateand safety profile, enrollment in phase 2 began.

The primary safety endpoint of the study was the percentage of patients in phases 1 and 2 presenting with severe AEs at 24 hours (immediate) and 1 month (short-term) post-endovascular procedure.

Venous outflow restoration outcomes

The primary hemodynamic endpoint over the follow-up period was the success of the endovascular treatment to restore venous outflow by >75% at month 1 compared to baseline, as measured by venous hemodynamic insufficiency severity score (VHISS) on duplex examination.5 VHISS is an ordinal scale of the overall extent and number of venous hemodynamic (VH) flow pattern anomalies, with a higher value of VHISS indicating a greater severity of VH flow pattern anomalies. The minimum possible VHISS value is 0; the maximum is 16.

We also reassessed extracranial and transcranial duplex VH criteria1at 1, 3, and 6 months. A subject was considered CCSVI-positive at any time during follow-up if ≥2 CCSVI VH criteria in phase I or if ≥2 CCSVI extracranial VH criteria in phase II were fulfilled.

Clinical outcomes

The primary clinical endpoint was number of relapses over the 6 months between the two treatment arms in phase 2. Secondary clinical endpoints included efficacy between the two treatment arms in phase II based on changes in Expanded Disability Status Scale (EDSS),6 Multiple Sclerosis Functional Composite (MSFC),7 and 6-minutes walked distance at 1, 3, and 6 months compared to baseline. Investigators performing clinical assessments were blinded to treatment status.

MRI outcomes

All subjects were examined in a 3T GE Signa Excite HD 12.0 HDx scanner (General Electric, Milwaukee, WI) at baseline, 1, 3, and 6 months of the study. The following sequences were acquired: 2-dimensional (2D) multiplanar dual fast spin-echo (FSE) proton density and T2-weighted image (WI); fluid-attenuated inversion-recovery (FLAIR); 3-dimensional (3D) high-resolution (HIRES) T1-WI using a fast spoiled gradient echo (FSPGR) with magnetization-prepared inversion recovery (IR) pulse and spin echo (SE) T1-WI. Images were acquired with and without use of a single-dose intravenous bolus of 0.1 mMol/Kg Gd-DTPA 5 min after injection. Investigators performing image analyses were blinded to subject characteristics and clinical and treatment status. No scans were rejected due to suboptimal quality.

Lesion measures: New lesion activity and proportion of active scans were the primary MRI endpoints of the study. The T2-, T1-, and contrast-enhancing (CE) lesion number and lesion volumes were measured on FLAIR and on T1 pre- and post-contrast images, respectively, using a semi-automated edge detection contouring/thresholding technique.8 To improve accuracy, all relevant within-subject images were co-registered to baseline.9All subsequent lesion analysis was done using the co-registered images. FLAIR and T1 post-contrast images were the primary sources, along with supporting images including proton density, T2, and longitudinal subtraction images (produced via voxel-wise subtraction of the previous time-point). For 6- and 12-month analyses, previous cross-sectional regions of interest were overlaid to facilitate the identification of new and newly enlarging T2 lesions. The cumulative number of T2, T1 and CE lesions was obtained by summing the total number of these lesions between all timepoints of the study.

Brain volume measures: For baseline analyses, FMRIB’s Structural Image Evaluation, using Normalisation of Atrophy, cross-sectional method (SIENAX) software was used (version 2.6),10 with corrections made for T1-hypointensity misclassification using an in-house developed in-painting program on T1-WI 3D images.11For longitudinal changes of the whole brain volume, we applied the SIENA method to calculate the percentage of brain volume change over 6 months.10 To quantify longitudinal gray matter and white matter percentage volume changes, we used a modified hybrid of FMRIB’s SIENA and SIENAX methods. We used a brain- and skull-constrained co-registration technique to place both baseline and follow-up images into a joint space halfway between the two at all timepoints in the study. Next, we combined baseline and follow-up intracranial volume masks via union and valid voxel masks via intersection, ensuring that the same imaging volume was analyzed at both timepoints. Finally, we segmented the resulting images with a modified longitudinal version of FMRIB’s Automated Segmentation Tool12 that used a 4-dimensional joint hidden random Markov field to prevent misclassification between timepoints when longitudinal intensity changes are lacking (or minimal). Total tissue volume was calculated for both baseline and 6-month follow-up for each tissue compartment from partial volume maps and percentage volume change was derived directly from the images.

Cognitive and quality of life (QoL) outcomes

Tertiary endpoints of the study included evaluation of patient self-reported QoL as measured by general QoL questionnaires (MSQoL-54),13 specific MS fatigue questionnaires [Fatigue Severity Scale (FSS)],14 the general patient impression of status change including Beck Depression Inventory Fast Screen (BDIFS)15 and MS Neuropsychological Screening Questionnaire (MSNQ),16 as well as neuropsychological status [Symbol Digit Modality Test (SDMT)],17 at baseline, 1, 3, and 6 months. The SDMT was applied only in phase 2.

Figure e-1

CONSORT Flow Diagram: PREMiSe Phase II

Table e-1

Baseline MRI characteristics of patients enrolled in the PREMiSe study.

Phase 1
(n=10) / Phase 2
Sham arm
(n=10) / Phase 2
Treatment arm
(n=9) / p
value*
Number of T2 lesions, mean (SD) median (min / max) / 28.1 (22.6) 23.5
(2 / 67) / 25.5 (16.9) 23
(5 / 54) / 25.9 (23.8) 18
(6 / 80) / 0.968
T2-LV, mean (SD) median (min / max) / 4.2 (4.7) 1.9
(0.2 / 14.3) / 15.1 (27.2) 5.1 (0.3 / 88.1) / 4.2 (5.7) 2.2 (0.1 / 16.9) / 0.256
Number of T1 lesions, mean (SD) median (min / max) / 11.9 (15.9) 5.5 (0-50) / 14.6 (14.3) 16.5 (0-49) / 9.2 (13) 3
(0-37) / 0.401
T1-LV, mean (SD) median (min / max) / 1 (1.1) 0.5
(0 / 2.9) / 4.2 (7.2) 0.9
(0 / 21.9) / 1.2 (2) 0.4
(0 / 5.6) / 0.246
Number of CE lesions, mean (SD) median (min / max) / 0.2 (0.4) 0
(0 / 1) / 0.3 (0.7) 0
(0 / 2) / 0.8 (0.9) 0.5
(0 / 2) / 0.238
CE-LV, mean (SD) median (min / max) / 0.04 (0.1) 0
(0 / 0.4) / 0.05 (0.1) 0
(0 / 0.4) / 0.05 (0.08) 0.1 (0 / 0.2) / 0.997
Normalized WBV, mean (SD) median (min / max) / 1477.1 (74.1) 1460
(1368.8 / 1604.3) / 1475 (102.5) 1511.8
(1245.2 / 1605.9) / 1474.8 (52.3) 1485.1
(1400 / 1552) / 0.997
Normalized GMV, mean (SD) median (min / max) / 741.9 (46.9) 756.2
(656.8-798) / 737.1 (54.3) 745.1
(628.7 / 798) / 736.4 (33) 729
(684.3 / 787.5) / 0.977
Normalized WMV, mean (SD) median (min / max) / 735.2 (45.9) 712
(697 / 827) / 737.5 (51.8) 754.3
(616.5 / 782.7) / 738.3 (31.5) 741.3
(697 / 774.8) / 0.971

Abbreviations: PREMiSe=Prospective Randomized Endovascular therapy in Multiple Sclerosis; SD=standard deviation; LV=lesion volume (reported in ml); CE=contrast-enhancing; WBV=whole brain volume; GMV=gray matter volume; WMV=white matter volume

* p value represents statistical analysis between sham-treated and angioplasty-treated arms of phase 2. The analysis between the treatment groups was performed by using Student's t-test.

Lesion volume data are presented in milliliters.

Table e-2

Baseline cognitive and quality of life characteristics of patients enrolled in the PREMiSe study.

Phase 1
(n=10) / Phase 2
Sham arm
(n=10) / Phase 2
Treatment arm
(n=9) / p
value*
SDMT, mean (SD) median (min / max) / NA / 54.9 (9.2) 51
(41 / 67) / 47.6 (16) 50
(28 / 78) / 0.278
BDIFS, mean (SD) median (min / max) / 1.9 (1.7) 2
(0 / 5) / 4 (2.4) 4
(1 / 8) / 4 (2.5) 4
(1 / 7 / 0.968
FSS, mean (SD) median (min / max) / 43.3 (13.1) 46
(21 / 63) / 48.1 (12.2) 50
(15 / 80) / 44.8 (13.7) 50
(20 / 57) / 0.842
MSNQ, mean (SD) median (min-max) / 18.2 (9.1) 19
(3 / 38) / 25.1 (12.3) 24.5
(5 / 48) / 22.2 (10.5) 25
(26.7 / 78.1) / 0.905
MSQoL-54 physical health composite, mean (SD) median (min / max) / 57.2 (15.6) 54.1
(37.7 / 85.3) / 55.5 (14.2) 53.5
(35.1 / 75.7) / 48.9 (15.3) 49.3
(26.7 / 78.1) / 0.400
MSQoL-54 mental health composite, mean (SD) median (min / max) / 71.2 (21.6) 73.4
(35.2 / 97.4) / 67.3 (16.4) 66.9
(46.1 / 92.7) / 54.9 (15.6) 56
(34 / 86.6) / 0.113

Abbreviations: PREMiSe=Prospective Randomized Endovascular therapy in Multiple Sclerosis; SD=standard deviation; NA=not available; SDMT=Symbol Digit Modality Test; BVMT-R=Brief Visual Memory Test Revised; BDIFS=Beck Depression Inventory Fast Screen; FSS=Fatigue Severity Scale; MSNQ=Multiple Sclerosis Neuropsychological Screening Questionnaire; MSQoL-54=Multiple Sclerosis Quality of Life 54 questionnaire

*p value represents statistical analysis between phase 2 arms. Analysis between treatment groups was performed by using Mann-Whitney rank sum test.

Siddiqui and Zivadinov, Supplement Page 1

Figure e-2: Numbers at risk for relapse, as shown by Kaplan-Meier plot for sham (group 1) and treatment (group 2) arms

Figure e-3:

Expanded Disability Status Scale (EDSS) (left), Multiple Sclerosis Functional Composite (MSFC) (middle), and 6 minutes walked distance in feet (right) values at baseline, 1, 3, and 6 months in phases 1 and 2 by using mixed-effect model analysis. P-values in the plots are based on comparison between the phase 2 treatment groups. Time effect p-values within treatment groups are for EDSS: phase 1 (p=0.386), phase 2 treated arm (p=0.525), and phase 2 sham arm (p=0.114), respectively; for MSFC: phase 1 (p=0.01), phase 2 treated arm (p=0.557), and phase 2 sham arm (p=0.04), respectively; and for 6 minutes walked distance: phase 1 (p=0.09), phase 2 treated arm (p=0.254), and phase 2 sham arm (p=0.257), respectively.

Figure e-4:

Graphs plotting Symbol Digit Modality Test (SDMT) (upper left), Beck Depression Inventory Fast Screen (BDIFS) (upper middle), Fatigue Severity Scale (FSS) (upper right), Multiple Sclerosis Neuropsychological Screening Questionnaire (MSNQ) (lower left), Multiple Sclerosis Quality of Life 54 (MSQoL-54) physical (lower middle) and mental health composite (lower right) questionnaire values at 0, 1, 3, and 6 months in phases I and II. P-values are based on comparison between phase II treatment groups by using mixed-effect model analysis. Time effect p-values within treatment groups are for SDMT: phase II treated arm (p=0.009), and phase II sham arm (p=0.392), respectively; for BDIFS: phase I (p=0.718), phase II treated arm (p=0.207), and phase II sham arm (p=0.01), respectively; for FSS: phase 1 (p=0.08), phase II treated arm (p=0.543), and phase II sham arm (p=0.03), respectively; for MSNQ: phase I (p=0.09), phase II treated arm (p=0.850), and phase II sham arm (p=0.004), respectively; for MSQoL-54 physical: phase I (p=0.02), phase II treated arm (p=0.0008), and phase II sham arm (p=0.001), respectively; and for MSQoL-54 mental: phase I (p=0.09), phase II treated arm (p=0.003), and phase II sham arm (p=0.07), respectively.

Siddiqui and Zivadinov, Supplement Page 1

References

1.Zivadinov R, Marr K, Cutter G, et al. Prevalence, sensitivity, and specificity of chronic cerebrospinal venous insufficiency in MS. Neurology 2011;77:138-144.

2.Dolic K, Marr K, Valnarov V, et al. Sensitivity and specificity for screening of chronic cerebrospinal venous insufficiency using a multimodal non-invasive imaging approach in patients with multiple sclerosis. Funct Neurol 2011;26:205-214.

3.Dolic K, Marr K, Valnarov V, et al. Intra- and extraluminal structural and functional venous anomalies in multiple sclerosis, as evidenced by 2 noninvasive imaging techniques. AJNR Am J Neuroradiol 2012;33:16-23.

4.Karmon Y, Zivadinov R, Weinstock-Guttman B, et al. Comparison of intravascular ultrasound with conventional venography for detection of extracranial venous abnormalities indicative of chronic cerebrospinal venous insufficiency. J Vasc Interv Radiol 2013;24:1487-1498.

5.Zamboni P, Menegatti E, Weinstock-Guttman B, et al. The severity of chronic cerebrospinal venous insufficiency in patients with multiple sclerosis is related to altered cerebrospinal fluid dynamics. Funct Neurol 2009;24:133-138.

6.Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33:1444-1452.

7.Cutter GR, Baier ML, Rudick RA, et al. Development of a multiple sclerosis functional composite as a clinical trial outcome measure. Brain 1999;122 ( Pt 5):871-882.

8.Zivadinov R, Rudick RA, De Masi R, et al. Effects of IV methylprednisolone on brain atrophy in relapsing-remitting MS. Neurology 2001;57:1239-1247.

9.Jenkinson M, Bannister P, Brady M, Smith S. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 2002;17:825-841.

10.Smith SM, Zhang Y, Jenkinson M, et al. Accurate, robust, and automated longitudinal and cross-sectional brain change analysis. Neuroimage 2002;17:479-489.

11.Zivadinov R, Heininen-Brown M, Schirda CV, et al. Abnormal subcortical deep-gray matter susceptibility-weighted imaging filtered phase measurements in patients with multiple sclerosis: a case-control study. Neuroimage 2012;59:331-339.