Recommended standards for reports dealing with arteriovenous hemodialysis accesses☆☆☆

Article Outline

• Abstract
• Classification of indications for access placement
  • Temporary hemodialysis
  • Long-term hemodialysis
• Characterization system and nomenclature
  • Conduit
  • Location
  • Configuration
  • Categories for catheters
• Preoperative evaluation
  • Grading of factors that modify outcome
  • Grading of anatomic factors that affect access patency
  • Arterial inflow
    • Pulse examination
    • Segmental blood pressures
    • The Allen Test
    • Arterial and arteriovenous access measurements with duplex ultrasound scanning
    • Conventional arteriography or magnetic resonance angiography
  • Venous outflow
    • Clinical examination
    • Duplex ultrasound scanning in the assessment of the venous circuit
      • Superficial veins
      • Deep veins
    • Contrast venography
• Grading of complications
  • Steal syndrome
  • Venous hypertension
  • Neuropathy
• Patency assessment
  • Postintervention patency
  • Functionality
  • Intent to treat10
  • Revision versus loss of arteriovenous access
  • Arteriovenous access removal or ligation because of reasons other than thrombosis
  • Time of measurement of patency
  • Definitions of arteriovenous access patency
    • Primary patency
    • Assisted primary patency
    • Secondary patency
  • Definitions of postintervention patency
    • Surgical interventions
    • Endovascular interventions
    • Definitions of postintervention patency
      • Postintervention primary patency
      • Postintervention assisted primary patency
      • Postintervention secondary patency
  • Definitions of catheter patency
    • Primary catheter site patency
    • Assisted primary catheter site patency
    • Secondary catheter site patency
  • Access vein survival
  • Estimating patency rates
• References
• Copyright

Abstract 

The incidence rate of treated end-stage renal disease in the united states is 180 per million and continues to rise at a rate of 7.8% per year. Arteriovenous hemodialysis access (AV access) creation and maintenance are two of the most difficult issues associated with the management of patients on hemodialysis. The 1-year complication rate of a primary prosthetic AV access for hemodialysis ranges from 33% to 99%. Various investigators report on patency and complications of AV access. However, it is rather difficult to compare outcomes because of the wide variety of access materials, configurations, locations, risk factors, and quality of inflow and outflow vessels. Although there have been reporting standards for dialysis access endovascular interventions and for central venous access placement, standards regarding surgical access placement and its revision are lacking. The “Dialysis Outcome Quality Initiative,” published by the National Kidney Foundation, provides recommendations for optimal clinical practices aimed at improving dialysis outcome and patient survival. This reporting standards document is not meant to be a “practice guidelines” or “best practices” document. Rather, the purpose of this document is to provide standardized definitions related to AV access procedures and to recommend reporting standards for patency and complications, to be used by surgeons, nephrologists, and interventional radiologists, that will permit meaningful comparisons among AV access procedures. The terms, definitions, and categories featured in this article have been approved by the Committee on Reporting Standards of the Society for Vascular Surgery and the American Association for Vascular Surgery and should be observed in preparing manuscripts on AV accesses for submission to the Journal Of Vascular Surgery. (J Vasc Surg 2002;35:603-10.)

Back to Article Outline

Classification of indications for access placement 

Outcomes from published arteriovenous hemodialysis (AV) access experiences cannot be meaningfully compared unless a subanalysis of the relative indications used is included. Therefore, this is a recommended standard, and the following classification is submitted for that purpose.

Temporary hemodialysis 

Tunneled or nontunneled hemodialysis catheters are commonly used to provide temporary access in the following situations: 1, acute renal failure; 2, overdose or intoxication; 3, urgent access in patients with chronic renal failure without available mature access; 4, temporary access until functionality of AV access is restored or the access replaced; 5, as substitute access in patients with peritoneal dialysis while their abdomens are being “rested” before new peritoneal catheter placement (ie, after peritonitis severe enough to require peritoneal dialysis catheter removal); and 6, severe rejection episodes in transplant recipients (who have lost or not yet had AV access).

Long-term hemodialysis 

Patients undergo long-term hemodialysis for the following indications: 1, long-term treatment of chronic renal failure; and 2, awaiting kidney transplantation.

Back to Article Outline

Characterization system and nomenclature 

Many terms are used inconsistently to describe various configurations of AV access. A component-based system is recommended. This system is based on three essential categories to be included in describing each procedure: conduit, location, and configuration. Although the use of additional modifiers is not precluded by this system, they should be supplemental to the three essential components, described subsequently.

Conduit 

Autogenous is used to describe native vein, whether it is in situ, transposed, or translocated. An autogenous AV access is an access created by a connection between an artery and a vein whereby the vein serves as an accessible conduit. Nonautogenous AV access is an access created by connecting an artery to a vein with a graft. Nonautogenous can be divided into prosthetic, such as expanded polytetrafluoroethylene or Dacron, and biograft , such as bovine heterograft or human umbilical vein. Additional prosthetic descriptors, such as tapered, ringed, or thin walled, may be used as indicated.

Location 

The specific anatomic sites of origination and termination of the AV access should be included in this component. Arterial inflow site is reported first, followed by a hyphen, and then the venous outflow site. In instances in which such a descriptor may be ambiguous, the addition of a broader anatomic reference should be included. This is typically reported as the body region or area where the access procedure is located and where cannulation will occur. Common examples include forearm and upper arm. Less common examples include inguinal and body wall. (Eg, a brachial-cephalic access should be described as either a forearm brachial-cephalic or upper arm brachial-cephalic access.)

Configuration 

This component gives descriptive information regarding the anastomotic connection and course of the conduit. Initial descriptors include either direct or indirect. Additional descriptors, such as transposed, translocated, straight, or looped, may be used. A direct access refers to a connection between native artery and vein and includes end-to-side, side-to-side, and end-to-end anastomoses. An indirect access is one in which either an autogenous or prosthetic material is inserted between the artery and vein to establish the connection. Because all prosthetic access materials by definition are of the indirect configuration, the descriptor indirect may be omitted when a prosthetic access is described.

An access performed with a transposed vein is a transposition and is a subtype of the direct descriptor and may be used in its place. It is used when the peripheral portion of the vein is moved from its original position, typically through a more superficial tunnel and connected to the artery, and the more central portion remains intact in its native location. The much less commonly used translocation access is a subtype of the indirect descriptor and may be used in its place. It describes a vein that has been disconnected both proximally and distally and is inserted in a position remote from its origin requiring anastomoses to both the arterial and venous segments of the access. Looped and straight refer to the course of the conduit. It should be noted that their inclusion is not essential when the configuration can be inferred from the anatomic descriptors. Tables I to IV outline the newly recommended nomenclature.

Table I. Forearm access procedures

ePTFE, Expanded polytetrafluoroethylene.

Table II. Upper arm access procedures

Table III. Lower extremity access procedures

ePTFE, Expanded polytetrafluoroethylene.

Table IV. Body wall access procedures

ePTFE, Expanded polytetrafluoroethylene.

Categories for catheters 

Catheters have traditionally been referred to as either temporary or permanent. These descriptors are often inaccurate, and catheters intended for temporary access and designed to be placed percutaneously without the need for a subcutaneous tunnel should be described as nontunneled catheters followed by their vein of insertion. Catheters intended for prolonged usage and placed through a subcutaneous tunnel should be described as tunneled followed by their vein of insertion. Although discouraged, catheters can also be referred to as cuffed and noncuffed to indicate whether they are equipped with a subcutaneous cuff to promote tissue ingrowth and further fixation of the catheter. Examples of possible dialysis catheter descriptors are: tunneled right internal jugular vein hemodialysis catheter, nontunneled left internal jugular vein hemodialysis catheter, tunneled right femoral vein hemodialysis catheter, and tunneled left femoral vein hemodialysis catheter.

Back to Article Outline

Preoperative evaluation 

Grading of factors that modify outcome 

It is difficult to compare outcomes among reported studies if factors that affect the results of particular access procedures are not identified and graded. A suggested grading scheme is presented in Table V. The grading scheme is organized into systemic factors, local limb factors, and miscellaneous factors that may affect outcome. The timeliness of placement and the material of the AV access are important factors in providing hemodialysis. Other factors that may affect and modify outcomes are also outlined in Table V, and a grading system is recommended for each.

Table V. Grading of factors that affect outcome

IVDA, Intravenous drug abuse; AV, arteriovenous.

Grading of anatomic factors that affect access patency 

The anatomy of the arterial and venous circuits involved in the placement of the AV access should be evaluated and findings reported so results can be evaluated in a meaningful way. If clinical examination results are clearly satisfactory, the access is placed. If not, noninvasive and even invasive methods are used for evaluation.¹, ² The following discussion identifies recommended methods for grading findings of diagnostic evaluations for reporting purposes.

Arterial inflow 

The axillary, brachial, radial, and ulnar artery pulses are carefully palpated in both upper extremities, and, when indicated, the femoral, popliteal, and pedal arteries are palpated. We propose the following scoring system: healthy pulse will be given a score of “0,” a diminished pulse a score of “1,” and an absent pulse a score of “2.” Clearly, a subcritical inflow lesion may be missed in the preoperative pulse evaluation. One should suspect this condition in patients who have had radiation therapy in the vicinity of the axillary and brachial arteries and in patients with multiple arterial occlusive lesions elsewhere.³

The difference in blood pressure between the two extremities should be reported and graded with the following scoring system: no difference will be given a score of “0,” a less than 10 mm Hg difference a score of “1,” a difference between 10 and 20 mm Hg a score of “2,” and a difference of more than 20 mm Hg a score of “3.”

The Allen Test⁴ is used to assess the arterial competence of the palmar arch. Definitions of positive and negative Allen test results for each of the radial and ulnar arteries are seen in the following description. Compression of both radial and ulnar arteries is used while the fist is clenched, then the fist is relaxed revealing blanched palm. For the test results to be defined as positive for radial artery insufficiency, the blanching continues 5 seconds or more after release of radial artery compression while the ulnar artery compression continues. For the test results to be defined as positive for ulnar artery insufficiency, blanching continues 5 seconds or more after release of ulnar artery compression while the radial artery compression continues.

Duplex ultrasound scan can determine the diameter of the artery. It has been found that a preoperative radial artery diameter of less than 1.6 mm resulted in failure of the radial-cephalic wrist autogenous AV access.⁵ In addition, duplex ultrasound scan can determine the velocity of blood flow in the AV access after construction. This measurement has been found to influence the long-term patency of the radial-cephalic wrist autogenous AV access. A flow rate of greater than 400 mL/min two weeks after access creation indicated a successful access, whereas a flow rate of less than 250 mL/min two weeks afterward indicated an access destined to fail.⁵

Contrast arteriography remains the gold standard for the evaluation of a suspected inflow stenosis or occlusion. When in doubt regarding the adequacy of the donor artery or the runoff, it is advisable to obtain an arteriogram that shows the entire arterial system from the origin of the subclavian to the distal branches. Magnetic resonance angiography can also be used for the same purpose.

Venous outflow 

Gross assessment of the superficial veins may be accomplished by first placing the arm in a dependent position and waiting for the veins to fill up naturally. Visual enhancement of the superficial veins can be provoked by placing a tourniquet on the upper arm while the patient clenches and releases the ipsilateral hand several times. On natural or provoked vein distention, one should measure the outside diameter of the vein. The following grading system is used to score the vein diameter: a vein that is more than 5 mm is scored as “0,” 4 to 5 mm scored as “1,” 3 to 4 mm scored as “2,” and less than 3 mm scored as “3.”

Superficial veins 

The adequacy of the superficial veins can be determined with duplex ultrasound scan. Examinations are performed with a tourniquet placed on the patient's mid forearm, followed by placement of the tourniquet on the upper arm. The basilic and cephalic veins are imaged with the highest frequency available (10 to 12 MHz) to ensure an acceptable B-mode imaging. Details of the venous lumen, such as webs, can be visualized and should be noted if present. On the basis of its lumen, a vein will be scored as follows: healthy lumen and wall scored as “0,” the presence of webs or sclerosis with fixed stenosis scored as “1,” and a vein that is occluded is scored as “2.” In addition, the length of usable vein should be measured, and the length is scored as follows: a vein length of more than 14 cm will be given the score of “0,” 7 to 14 cm a score of “1,” and less than 7 cm a score of “2.”

Deep veins 

Duplex scanning of the deep venous system for evaluation of stenosis or occlusion may be performed. The quality of the deep vein lumen is scored the same way as the superficial vein is scored.

Venography may also be indicated to determine the patency and adequacy of the superficial and deep venous systems, including central veins.

Back to Article Outline

Grading of complications 

Characteristics and grading of complications are detailed in Table VI.

Table VI. Grading severity of arteriovenous-access complications

AV, Arteriovenous.

Steal syndrome 

All AV accesses divert or “steal” blood from the distal circulation.⁶ A clinical syndrome resulting from this loss of distal circulation occurs when the various local compensatory mechanisms fail. This creates a zone of arterial insufficiency in the tissues distal to the fistula with all the usual clinical manifestations of vascular insufficiency: pain, ischemic neuropathy, ulceration, and gangrene.

Venous hypertension 

Venous hypertension invariably occurs with the arterialization of the venous system. Clinical problems may be manifested with severe degrees of venous hypertension.⁷

Neuropathy 

Neuropathy in the patient for hemodialysis with a fistula should always raise the suspicion of underlying ischemia as the result of a steal syndrome. It is by far the most common cause of a unilateral neuropathy in these patients. Only when this diagnosis has been carefully excluded should other causes be considered. These include uremic neuropathy, diabetic neuropathy, carpal tunnel syndrome, and other compartment syndromes, such as the cubital or ulnar nerve compression syndrome.⁸

Back to Article Outline

Patency assessment 

Valid comparisons on patency can be made only if patency is defined in a way that can be universally used by all specialties in a consistent manner. Before defining the various types of patency, there are several important issues to be considered:

Postintervention patency 

To have definitions of patency that are valid and complete, it becomes apparent that not only should patency from the time of initial access placement be defined but also the patency of the subsequent interventions. A radiologist or a surgeon who is interested in comparing the ability of different methods in maintaining or reestablishing patency of an access should be able to measure and evaluate interval patency since that intervention, not having to worry about the overall (life-span) patency of an access.

Functionality 

Successful hemodialysis is reliant on the functionality of AV access. The word “functional” can be added to qualify the word “patency.” Thus, “primary access patency” can be modified to “primary access functional patency” to emphasize the functionality of the access and its continuous use for successful hemodialysis during its indicated patency. An access is said to be “functional” when it is able to deliver a flow rate of 350 to 400 mL/min without access recirculation to maintain a treatment time less than 4 hours.⁹ A nonfunctional AV access is an access that is not being successfully used for hemodialysis whether it is patent or not. However, a failing AV access is an access that is patent, not thrombosed, but in which thrombosis and failure seems imminent because of hemodynamic impairments.

Intent to treat¹⁰ 

It is not uncommon for authors to acknowledge an initial failure rate for a procedure but then exclude these failures from the final calculation of cumulative patency rates. Although information is not withheld, the practice is deceiving. For example, if 25% of a particular intervention initially failed but 80% of those successful were still patent after a given time period, only 60% of those initially treated will have benefitted. This “intent to treat” rule may occasionally seem too strict, as in a patient in whom there is inability to gain percutaneous access for an endovascular procedure or in a patient in whom a planned outflow revision with patch angioplasty is not feasible, and instead a new access is placed. However, to arbitrarily allow such “reasonable” variances creates more problems than it solves, and therefore it is advised that the “intent to treat” rule be used in all reports dealing with patency of AV access.

Revision versus loss of arteriovenous access 

Revision of an AV access may require segmental replacement. If the remaining segment can be used for dialysis while the new segment is incorporating or maturing, the access is considered revised and not replaced. However, if the remaining segment is not adequate for continuous dialysis and the patient needs a temporary catheter until the newly placed segment matures, then the access is considered replaced.

Arteriovenous access removal or ligation because of reasons other than thrombosis 

If an AV access is removed, ligated, or abandoned because of reasons other than thrombosis (ie, infection, steal, venous hypertension, or successful transplantation), then access functionality is considered to have ceased, and patency is calculated as continuing up to the time of permanent cessation of dialysis using the access.

Time of measurement of patency 

The time of measurement of patency is the point in time when observation or censoring of an access is performed.

Definitions of arteriovenous access patency 

This is the interval from the time of access placement until any intervention designed to maintain or reestablish patency, access thrombosis, or the time of measurement of patency.

This is the interval from the time of access placement until access thrombosis or the time of measurement of patency, including intervening manipulations (surgical or endovascular interventions) designed to maintain the functionality of a patent access (number of interventions can be represented numerically in between brackets after the actual patency of an access; ie, assisted primary patency of 18 months [2]. This indicates that the access has been functional for 18 months but needed two interventions to maintain its functionality).

This is the interval from the time of access placement until access abandonment, thrombosis, or the time of patency measurement including intervening manipulations (surgical or endovascular interventions) designed to reestablish functionality in thrombosed access.

Definitions of postintervention patency 

Before defining postintervention patencies, we will define and describe various interventions. An intervention is any surgical or endovascular manipulation of the access to maintain or reestablish its functionality. Although the following list may change with the addition of new methods and technologies, such interventions can be divided into surgical and endovascular.

A, Thrombectomy: surgical removal of thrombus from the AV access; B, outflow patch angioplasty: surgical revision of the venous outflow stenosis with a prosthetic or autogenous patch to enlarge a stenosed outflow anastomosis; C, primary repair: a stenotic area is resected and an end-to-end anastomosis is performed; and D, segmental access replacement: this can be one of four types: 1, venous outflow: surgical replacement of the distal segment of the access to bypass an outflow venous stenosis; 2, mid-access: surgical replacement of segment of the access not inclusive of its inflow or outflow anastomoses; 3, arterial inflow: surgical replacement of the proximal segment of the access to bypass an inflow arterial stenosis to a different (usually more proximal) arterial inflow; and 4, any combination of the previous can be used such as venous outflow and mid-access replacement.

A, Balloon angioplasty: intraluminal balloon dilatation; B, stent deployment: placement of a self-expanding or balloon expandable stent; C, pharmacologic thrombolysis: catheter-directed infusion of thrombolytic agents; D, mechanical thrombolysis and thrombectomy: fragmentation, maceration, or mobilization of thrombus by mechanical means or devices; and E, pharmacomechanical thrombolysis: catheter-directed administration of an agent that results in pharmacologic thrombolysis accompanied by mechanical disruption of thrombus by fragmentation, maceration, or mobilization.

The following definitions of postintervention patencies differ slightly but are consistent with those published by the Society for Cardiovascular and Interventional Radiology (SCVIR).¹¹ Of note, the SCVIR document applies only to percutaneous endovascular manipulations. This document's definition of intervention encompasses all surgical and endovascular manipulations.

Postintervention primary patency 

This is the interval from the time of any surgical or endovascular intervention until thrombosis or the time of measurement of patency.

Postintervention assisted primary patency 

This is the interval from the time of intervention until first access thrombosis or the time of patency measurement, including all intervening similar manipulations (surgical or endovascular interventions) designed to maintain the functionality of the access. Of note, patency ends for a particular intervention if another method is used to maintain patency. For example, if a venous anastomotic lesion is dilated and recurs and a surgical jump graft is performed, the assisted primary patency of the balloon dilatation procedure has ended (the number of interventions can be represented numerically in between brackets after the actual patency of an access; ie, assisted postintervention primary patency of 6 months [1]. This indicates that this access has an uninterrupted postintervention patency of 6 months but needed one similar reintervention to maintain its functionality).

Postintervention secondary patency 

This is the interval from the time of intervention until access abandonment or the time of patency measurement, including similar intervening manipulations (surgical or endovascular interventions) designed to reestablish functionality in thrombosed access. The same definition is described as “postintervention access patency” in the SCVIR document.¹¹

Definitions of catheter patency 

It is important to distinguish catheter patency from the more important concept of catheter site patency because exchange for a new catheter over a wire would end catheter patency but maintain site patency. This concept has been identified previously, and we use it in a modified form for conformity because the original document emphasized endovascular interventions .¹² In addition, we introduce the concept of access vein survival to account for availability of a certain vein (internal jugular, femoral) for future access. We propose the following definitions.

This is the interval from time of placement of the catheter until first intervention, catheter malfunction, completion of therapy, and catheter site abandonment or time of measurement of patency.

This is the interval from time of placement until catheter malfunction, completion of therapy, and catheter site abandonment or time of measurement of patency including intervening manipulations (endovascular or surgical interventions) aiming to restore functionality of the catheter without the need for its replacement (number of manipulations can be represented numerically in between brackets after the actual patency of a catheter; ie, assisted primary catheter site patency of 2 months [2]. This indicates that this catheter has an assisted primary catheter site patency of 2 months but needed two interventions to restore its functionality).

This is the interval from time of placement until catheter site abandonment, completion of therapy, or time of measurement of patency including catheter replacements (exchanges) provided the access site is maintained. The number of catheter replacements while maintaining the same site should be indicated (number of manipulations can be represented numerically in between brackets after the actual patency of a catheter; ie, secondary patency of 2 months [2]. This indicates that this site has a secondary patency of 2 months but needed two catheter exchanges maintaining the original site).

Access vein survival 

The concept of access vein patency is an important one because a given vein can be used repeatedly throughout the patient's life on hemodialysis. Access vein survival is defined as the period from the time point when a vein (internal jugular, femoral, etc) was first used for access until the vein is no longer usable for that purpose.

Estimating patency rates 

Life table (LT) method¹³, ¹⁴ and Kaplan-Meier (KM) survival curves¹⁵ are acceptable methods to estimate patency. In figures, the numbers of patients at risk at the start of each interval (periodically for the KM) must be included and the standard error for each estimate of patency must be displayed with bars. When the standard error of the patency rate estimate exceeds 10%, the curve either should not be drawn or should be represented with a dotted line as a means of indicating lack of reliability of the estimate. Comparisons of patency curves should be performed with the log-rank test of significance.¹⁶ Complete LT and KM data should be submitted as an appendix with each report, to allow the validity of the data to be evaluated.

Separate LTs should be provided for each type of operative procedure. When pertinent, additional LT or KM plots should be reported for major risk or treatment factors that appear to affect patency. In some instances, interdependence of variables will limit the confidence with which conclusions may be drawn from this subgrouping of data. In this situation, it is often appropriate to apply multivariate analysis with the Cox hazard regression analysis.¹⁷

Back to Article Outline

References 

1. Rutherford RB, Fleming PW, McLeod FD. Vascular diagnostic methods for evaluating patients with arteriovenous fistula. In:  Diethrich ED editors. Noninvasive cardiovascular diagnosis: current concepts. Baltimore: University Park Press; 1978;p. 189–203
   • View In Article
2. Rutherford RB. Noninvasive testing in the diagnosis and assessment of arteriovenous fistula. In:  Bernstein EF editors. Noninvasive diagnostic techniques in vascular disease. St Louis: Mosby; 1982;p. 430–442
   • View In Article
3. Silva MBJ, Hobson RW, Pappas PJ, Jamil Z, Araki CT, Goldberg MC, et al.  A strategy for increasing use of autogenous hemodialysis access procedures: impact of preoperative noninvasive evaluation. J Vasc Surg. 1998;27:302–307
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (70 KB)
   • CrossRef
4. Allen EV. Thromboangitis obliterans: methods of chronic occlusive arterial lesions distal to the wrist with illustrative cases. [abstract] Am J Med Sci. 1929;178:273
   • View In Article
5. Wong V, Ward R, Taylor J, Selvakumar S, How TV, Bakran A. Factors associated with early failure of arteriovenous fistulae for haemodialysis access. Eur J Vasc Endovasc Surg. 1996;12:207–213
   • View In Article
   • Abstract
   • Full-Text PDF (661 KB)
   • CrossRef
6. DeCaprio JD, Valentine RJ, Kakish HB, Awad R, Hagino RT, Clagett GP. Steal syndrome complicating hemodialysis access. Cardiovasc Surg. 1997;5:648–653
   • View In Article
   • MEDLINE
   • CrossRef
7. Currier CBJ, Widder S, Ali A, Kuusisto E, Sidawy A. Surgical management of subclavian and axillary vein thrombosis in patients with a functioning arteriovenous fistula. Surgery. 1986;100:25–28
   • View In Article
   • MEDLINE
8. Konishiike T, Hashizume H, Nishida K, Inoue H, Moriwaki K. Cubital tunnel syndrome in a patient in long-term haemodialysis. J Hand Surg [Br]. 1994;19:636–637
   • View In Article
   • MEDLINE
   • CrossRef
9. Sands J, Jabayac P, Miranda C. Delivered blood flow in cuffed central venous dialysis catheters [abstract]. ASAIO J. 1997;43:69
   • View In Article
   • MEDLINE
   • CrossRef
10. Rutherford RB, Baker JD, Ernst C, Johnston KW, Porter JM, Ahn S, et al.  Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg. 1997;26:517–538
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1996 KB)
    • CrossRef
11. Gray RJ, Sacks D, Martin LG, Trerotola SO. Reporting standards for percutaneous interventions in dialysis access. Technology Assessment Committee. J Vasc Interv Radiol. 1999;10:1405–1415
    • View In Article
    • Full-Text PDF (1231 KB)
    • CrossRef
12. Silberzweig JE, Sacks D, Khorsandi AS, Bakal CW. Reporting standards for central venous access. Technology Assessment Committee J Vasc Interv Radiol. 2000;11:391–400
    • View In Article
    • Full Text
    • Full-Text PDF (89 KB)
    • CrossRef
13. Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, et al.  Design and analysis of randomized trials requiring prolonged observations of each patient. I. Introduction and design. Br J Cancer. 1976;34:585–612
    • View In Article
    • MEDLINE
    • CrossRef
14. Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, et al.  Design and analysis of randomized trials requiring prolonged observations of each patient. II. Analysis and examples. [abstract] Br J Cancer. 1977;35:1–39
    • View In Article
    • MEDLINE
    • CrossRef
15. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. [abstract] J Am Statist Assoc. 1958;8:699–712
    • View In Article
16. Lawless JF. Statistical models and methods for lifetime data. New York, NY: Wiley & Sons; 1982;
    • View In Article
17. Katz MH, Hauck WW. Proportional hazards (Cox) regression. J Gen Intern Med. 1993;8:702–711
    • View In Article
    • MEDLINE
    • CrossRef