Investigation Prediction of Arteriovenous Fistula Dysfunction: Can it be Taught? Joel E. Rosenberg, Alexander S. Yevzlin, Micah R. Chan, Amanda M. Valliant, and Brad C. Astor Division of Nephrology, Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin
ABSTRACT Physical examination (PE) is an excellent means of predicting arteriovenous fistula (AVF) dysfunction. Although quick and inexpensive, PE is seldom used as a tool to assess stenosis by general nephrologists, dialysis nurses, and dialysis technicians. Previous studies have demonstrated that PE can be taught to interventional specialists, but the perception remains that it is too complex to be performed by other health care professionals. We hypothesized that the physical exam can be taught to a nonmedical professional, and that, with time, it would be comparable to the physical exam performed by a full-time interventional specialist. An undergraduate student and an interventional specialist (MD) examined AVF for dysfunction in a tertiary care hospital over a 6month period. PE was performed on patients who were suspected of having dialysis access dysfunction and were referred for angiography and intervention (n = 49). Physical exam findings were categorized blindly by each examiner into four categories of lesion location: inflow,
outflow, both, or neither. Data were privately recorded and compared to the gold standard of angiographic results. Potential confounding variables, including age, gender, diabetic status, and location of AVF were recorded. Weighted Cohen’s kappa value was used as a measurement of the level of agreement beyond chance between the diagnoses made by physical exam and angiography. The full-time interventional specialist demonstrated correct prediction of lesion location of 89.8% (kappa = 0.850), while the undergraduate student had a correct prediction of 77.6% (kappa = 0.625). The student’s performance, however, differed significantly over time. The student correctly predicted the location of the lesion in 6 (42.9%) of the first 14 patients (kappa = 0.082), compared to 32 (91.4%) of the last 35 patients (kappa = 0.855). We suggest that physical exam of AVF can be taught to a nonmedical professional in a short duration of time and the predictive value of the exam can be similar to that of an interventional specialist.
Physical examination (PE) is an excellent means of predicting arteriovenous fistula (AVF) dysfunction due to ease of performance and cost effectiveness (1). It also has a relatively high diagnostic accuracy compared to the gold standard of angiography (2–6). Previous studies have demonstrated that PE can be taught to interventional specialists and nephrology fellows, but the perception remains that it is too complex to be performed by other health care professionals (1,4,5). Although quick and inexpensive, PE is seldom used in the hemodialysis unit as a means of predicting access stenosis by general nephrologists, dialysis nurses, and dialysis technicians, in part due to the perception that the physical exam is difficult to learn (2,7–10). We hypothesized that the physical exam can be taught
to a undergraduate student who is a nonmedical professional, and with time, the ability to accurately detect fistula stenosis would be comparable to the physical exam performed by a full-time interventional specialist. Methods An undergraduate student and an interventional specialist (MD) examined AVF for dysfunction in a tertiary care hospital over a 6-month period. A 1-day training was given to the student consisting of a complete walkthrough of the exam performed on three consecutive patients. Literature resources were given as well (2,8–11). Physical examination was performed on patients who were suspected of having dialysis access dysfunction and were referred for angiography and intervention (n = 49). Brief inspection of the arm, shoulder, breast, and neck for palpation and auscultation were performed systematically. Palpation was carefully performed from the fistula anastomosis all the way to the chest wall. Hyperpulsatile, normal and weak pulse were
Address correspondence to: Joel E. Rosenberg, Division of Nephrology, Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI, e-mail: [email protected]. Seminars in Dialysis—Vol 28, No 5 (September–October) 2015 pp. 544–547 DOI: 10.1111/sdi.12361 © 2015 Wiley Periodicals, Inc. 544
545
PHYSICAL EXAMINATION OF AVF
Statistical Evaluation
Percentage of Correct Predictions
Student Specialist
Number of Patients
Fig. 1. Correct prediction percentage by number of patients examined.
Accuracy of Physical Exam vs. Time
Month
Fig. 2. Correct prediction percentage by month.
Cohen’s kappa value was used to measure the level of agreement beyond chance between the diagnoses made by physical examination and the angiography in related studies (13,14). Results
Weighted Cohen’s kappa value analysis was performed using Stata, version 12.1 (www.Stata.com) to measure the level of agreement beyond chance between the dichotomous diagnoses (presence or absence of lesion) made by physical exam when compared to angiography. Weights of 0.5 were used if the examiner predicted just inflow or just outflow in a dysfunctional fistula that had both types of lesions (inflow lesion and outflow lesion) (Table 1).
TABLE 1. Layout of value given for Weighted kappa value. Weights of 0.5 were used if one of the lesions was detected but not the other in a fistula that had both inflow and outflow lesions
Outflow Inflow Both Neither
Prediction of Vascular Access Dysfunction
Percentage of Correct Predictions
assessed as well as thrill and bruit (continuous and discontinuous characteristics were noted). Pulse augmentation followed by arm elevation tests to evaluate inflow and outflow segments were performed in that order (9). An inflow segment was defined as any segment of vessel proximal to hemodialysis needle placement, where as an outflow segment was defined as any segment venous to where the needles for hemodialysis are placed. The diagnosis of an outflow lesion was based on the assessment of pulse and thrill to determine the presence of a water-hammer pulse (hyperpulsation), systolic thrill, and abnormal arm elevation test (8,9,12). The diagnosis of inflow lesion was based on the assessment of pulse and thrill to determine the presence of a weak pulse (hypopulsation and flat access), lack of a continuous thrill, and abnormal augmentation test (8,9,12). Clinical features that contribute to a central outflow stenosis also included edema of the arm and shoulder, breast, supraclavicular, neck, and face swelling. The augmentation test was considered normal if the fistula demonstrated augmentation of pulse upon occlusion of the outflow vessel distal to the fistula (9). The arm elevation test was performed by elevating the extremity with the fistula and examining the normal collapse of the fistula (12). An abnormal test was recorded if the fistula remained plump after arm elevation. The physical examination findings were categorized into four predictions of lesion location: inflow, outflow, both, or neither. Both the interventional specialist and student were masked as to the results of the other. Potential confounding variables, including age, gender, diabetic status, and location of AVF were recorded. The prediction was compared to the gold standard of AVF angiography. An independent third party reviewed the short video run of the angiograph and determined stenosis type.
Outflow
Inflow
Both
Neither
1.000 0.000 0.500 0.000
0.000 1.000 0.500 0.000
0.500 0.500 1.000 0.000
0.000 0.000 0.000 1.000
Forty-nine patients were examined in 6 months of patients referred with suspicion of fistula dysfunction. The average age of the subjects was 59.1 and majority of the subjects were male (61.2%), diabetic (53.1%), and had an upper arm AVF (75.5%). The full-time interventional specialist demonstrated correct prediction of lesion location of 89.8% (kappa = 0.850), while the undergraduate student had a correct prediction of 77.6% (kappa = 0.625). The student’s performance, however, differed significantly over time. The student correctly predicted the location of the lesion in 6 (42.9%) of the first 14 patients (kappa = 0.082), compared to 32 (91.4%) of the last 35 patients (kappa = 0.855). The interventional nephrologist correctly predicted the location of the lesion in 11 (78.6%) of the first 14 patients (kappa = 0.661), compared to 33 (94.3%) of the last 35 patients (kappa = 0.876).
546
Rosenberg et al.
TABLE 2. Weighted kappa values for physical examination by the undergraduate student and interventional nephrologist with expertise in physical examination Party Interventional specialist Undergraduate student
Correct diagnosis
Correct diagnosis in first 14 patients
Correct diagnosis in last 35 patients
91.9% kappa = 0.850 80.2% kappa = 0.625
78.6% kappa = 0.661 42.9% kappa = 0.082
94.3% kappa = 0.888 91.4% kappa = 0.830
There was substantial agreement beyond chance between the physical examination and the angiography in the diagnosis of outflow stenosis (85.7%, kappa = 0.697) by the undergraduate student and a moderate agreement beyond chance for prediction of inflow stenosis (79.6%, kappa = 0.583). The interventional nephrologist also had a substantial agreement beyond chance between the physical examination and the angiography in the diagnosis of outflow stenosis (93.9%, kappa = 0.863) and substantial agreement beyond chance in the diagnosis of inflow stenosis (81.6%, kappa = 0.874). When comparing lower arm fistulae in just the last 35 patients, the undergraduate student correctly predicted lesion location with a strong agreement beyond chance of 80.0% (kappa = 0.688), while the interventional specialist performed with a perfect agreement of 100.0% (kappa = 1.000). In upper arm fistulae, the student performed the exam with a higher level of accuracy than compared with the lower arm fistulae (93.33%, kappa = 0.854) and was in perfect agreement with the interventional nephrologist (93.33%, kappa = 0.854). Discussion This study demonstrates that an undergraduate student with no previous medical knowledge or physical examination experience can accurately predict lesion location in dysfunctional AVF at a comparable rate to that of a full-time interventional specialist (Figs. 1 and 2, Table 2). After examining 14 patients, the undergraduate’s ability to correctly diagnose lesion location was 91.4% for the subsequent 35 patients. The specialist’s ability to correctly predict fistula dysfunction by physical exam increased over time as well, suggesting performing exams will thus increase even a specialist’s ability to correctly predict fistula dysfunction. Correct prediction of outflow lesions was greater than that of inflow lesions. The reason for this is unclear, but may be related to the presence of other clinical features that provide diagnostic clues, including edema of the arm and shoulder, breast, supraclavicular region, neck and face, and presence of collateral vasculature on the chest wall. The results obtained support our hypothesis that the physical exam skill can be taught to a nonmedical professional in a relatively short amount of time. Although this study only examined location of stenosis in patients referred for intervention, detection of
stenosis in an unselected group may not be as easy but this information still proves useful. It follows that the exam of the vascular access can be taught to general nephrologists, dialysis technicians, and nurses that work in the hemodialysis unit. Teaching patients to assess their own fistula as well would also prove beneficial to increase access longevity by promptly reporting abnormal findings (15). While advanced technology such as ultrasound and static and dynamic venous pressure is often used, the physical exam is free and readily available and can be performed in less than 1 minute. Nephrology training centers should make the physical examination of vascular accesses an essential part of their training curriculum, as it would be extremely cost and time efficient. Future research should focus on educational interventions in the hemodialysis unit that promote appropriate and timely referral for angiography.
References 1. Leon C, Asif A: Physical examination of arteriovenous fistulae by a renal fellow: does it compare favorably to an experienced interventionalist? Semin Dial 21:557–560, 2008 2. Mishler R, Shon D, Hubert B, Nissenson AR: Development and usefulness of a physical examination toll to diagnose vascular access dysfunction. J Am Soc Nephrol 11:190A, 2000 3. Choi JR, Kim YS, Yoon SA, Won YD, Son YS, Song WJ, Song HC, Kim YS, Chang YS, Bang BK, Kim YO: Accuracy of physical examination in the detection of arteriovenous fistula dysfunction. Korean J Nephrol 25:797–802, 2006 4. Asif A, Cherla G, Merrill D, Cipleu CD, Briones P, Pennel P: Conversion of tunneled hemodialysis catheter-cosigned patients to arteriovenous fistula. Kidney Int 67:2399–2406, 2005 5. Asif A, Leon C, Orozco-Vargas LC, Krishnamurty G, Choi KL, Mercado C, Merrill D, Thomas I, Salman L, Artikov S, Bourgoignie JJ: Accuracy of physical examination in the detection of arteriovenous fistula stenosis. Clin J Am Soc Nephrol 2:1191–1194, 2007 6. Asif A, Leon C, Orozco-Vargas LC, Krishnamurty G, Choi KL, Mercado C, Merrill D, Thomas I, Salman L, Artikov S, Bourgoignie JJ: Accuracy of physical examination in the detection of arteriovenous graft stenosis. Semin Dial 21:85–88, 2008 7. National Kidney Foundation: K/DOQI clinical practice guidelines in vascular access: 2006 update. Am J Kidney Dis 48(Suppl. 1): S176–S306, 2006 8. Beathard GA: Physical examination of the dialysis vascular access. Semin Dial 11:231–236, 1998 9. Beathard GA: Physical examination: the forgotten tool. In: Gray R, Sands J (eds). A Multidisciplinary Approach for Hemodialysis Access. New York: Lippincott, Williams & Wilkins, 2002:11–118 10. Beathard GA: An algorithm for the physical examination of early fistua failure. Semin Dial 18:331–335, 2005 11. Beathard GA: Physical examination of the dialysis vascular access. Semin Dial 11:231–236, 1998 12. Beathard GA: A Practitioners Guide To Physical Examination of Dialysis Vascular Access. Dallas: The End Stage Renal Disease Network of Texas, Inc. 2003 13. McGinn T, Wyer PC, Newman TB, Keitz S, Leipzig R, Guyatt G: Tips for learns of evidence-based medicine: measures of observer variability (kappa statistic). CMAJ 171:1369–1373, 2004
PHYSICAL EXAMINATION OF AVF 14. Trerotola SO, Ponce P, Stavropoulos SW, Clark TWI, Tuite CM, Mondschein JI, Shlansky-Goldberg R, Freiman DB, Patel AA, Soulen MC, Cohen R, Alan Wasserstein A, Chittams JL: Physical examination versus normalized pressure ratio for prediction outcomes of hemodialysis access interventions. J Vasc Interv Radiol 14:1387–1393, 2003
547
15. Vassalotti JA, Jennings WC, Beathard GA, Neumann M, Caponi S, Fox CH, Spergel LM and the Fistula First Breakthrough Initiative Community Education Committee: Targeting Catheter Last in Fistula First. Semin Dial 25(3):303–310, 2012
ABSTRACT Physical examination (PE) is an excellent means of predicting arteriovenous fistula (AVF) dysfunction. Although quick and inexpensive, PE is seldom used as a tool to assess stenosis by general nephrologists, dialysis nurses, and dialysis technicians. Previous studies have demonstrated that PE can be taught to interventional specialists, but the perception remains that it is too complex to be performed by other health care professionals. We hypothesized that the physical exam can be taught to a nonmedical professional, and that, with time, it would be comparable to the physical exam performed by a full-time interventional specialist. An undergraduate student and an interventional specialist (MD) examined AVF for dysfunction in a tertiary care hospital over a 6month period. PE was performed on patients who were suspected of having dialysis access dysfunction and were referred for angiography and intervention (n = 49). Physical exam findings were categorized blindly by each examiner into four categories of lesion location: inflow,
outflow, both, or neither. Data were privately recorded and compared to the gold standard of angiographic results. Potential confounding variables, including age, gender, diabetic status, and location of AVF were recorded. Weighted Cohen’s kappa value was used as a measurement of the level of agreement beyond chance between the diagnoses made by physical exam and angiography. The full-time interventional specialist demonstrated correct prediction of lesion location of 89.8% (kappa = 0.850), while the undergraduate student had a correct prediction of 77.6% (kappa = 0.625). The student’s performance, however, differed significantly over time. The student correctly predicted the location of the lesion in 6 (42.9%) of the first 14 patients (kappa = 0.082), compared to 32 (91.4%) of the last 35 patients (kappa = 0.855). We suggest that physical exam of AVF can be taught to a nonmedical professional in a short duration of time and the predictive value of the exam can be similar to that of an interventional specialist.
Physical examination (PE) is an excellent means of predicting arteriovenous fistula (AVF) dysfunction due to ease of performance and cost effectiveness (1). It also has a relatively high diagnostic accuracy compared to the gold standard of angiography (2–6). Previous studies have demonstrated that PE can be taught to interventional specialists and nephrology fellows, but the perception remains that it is too complex to be performed by other health care professionals (1,4,5). Although quick and inexpensive, PE is seldom used in the hemodialysis unit as a means of predicting access stenosis by general nephrologists, dialysis nurses, and dialysis technicians, in part due to the perception that the physical exam is difficult to learn (2,7–10). We hypothesized that the physical exam can be taught
to a undergraduate student who is a nonmedical professional, and with time, the ability to accurately detect fistula stenosis would be comparable to the physical exam performed by a full-time interventional specialist. Methods An undergraduate student and an interventional specialist (MD) examined AVF for dysfunction in a tertiary care hospital over a 6-month period. A 1-day training was given to the student consisting of a complete walkthrough of the exam performed on three consecutive patients. Literature resources were given as well (2,8–11). Physical examination was performed on patients who were suspected of having dialysis access dysfunction and were referred for angiography and intervention (n = 49). Brief inspection of the arm, shoulder, breast, and neck for palpation and auscultation were performed systematically. Palpation was carefully performed from the fistula anastomosis all the way to the chest wall. Hyperpulsatile, normal and weak pulse were
Address correspondence to: Joel E. Rosenberg, Division of Nephrology, Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI, e-mail: [email protected]. Seminars in Dialysis—Vol 28, No 5 (September–October) 2015 pp. 544–547 DOI: 10.1111/sdi.12361 © 2015 Wiley Periodicals, Inc. 544
545
PHYSICAL EXAMINATION OF AVF
Statistical Evaluation
Percentage of Correct Predictions
Student Specialist
Number of Patients
Fig. 1. Correct prediction percentage by number of patients examined.
Accuracy of Physical Exam vs. Time
Month
Fig. 2. Correct prediction percentage by month.
Cohen’s kappa value was used to measure the level of agreement beyond chance between the diagnoses made by physical examination and the angiography in related studies (13,14). Results
Weighted Cohen’s kappa value analysis was performed using Stata, version 12.1 (www.Stata.com) to measure the level of agreement beyond chance between the dichotomous diagnoses (presence or absence of lesion) made by physical exam when compared to angiography. Weights of 0.5 were used if the examiner predicted just inflow or just outflow in a dysfunctional fistula that had both types of lesions (inflow lesion and outflow lesion) (Table 1).
TABLE 1. Layout of value given for Weighted kappa value. Weights of 0.5 were used if one of the lesions was detected but not the other in a fistula that had both inflow and outflow lesions
Outflow Inflow Both Neither
Prediction of Vascular Access Dysfunction
Percentage of Correct Predictions
assessed as well as thrill and bruit (continuous and discontinuous characteristics were noted). Pulse augmentation followed by arm elevation tests to evaluate inflow and outflow segments were performed in that order (9). An inflow segment was defined as any segment of vessel proximal to hemodialysis needle placement, where as an outflow segment was defined as any segment venous to where the needles for hemodialysis are placed. The diagnosis of an outflow lesion was based on the assessment of pulse and thrill to determine the presence of a water-hammer pulse (hyperpulsation), systolic thrill, and abnormal arm elevation test (8,9,12). The diagnosis of inflow lesion was based on the assessment of pulse and thrill to determine the presence of a weak pulse (hypopulsation and flat access), lack of a continuous thrill, and abnormal augmentation test (8,9,12). Clinical features that contribute to a central outflow stenosis also included edema of the arm and shoulder, breast, supraclavicular, neck, and face swelling. The augmentation test was considered normal if the fistula demonstrated augmentation of pulse upon occlusion of the outflow vessel distal to the fistula (9). The arm elevation test was performed by elevating the extremity with the fistula and examining the normal collapse of the fistula (12). An abnormal test was recorded if the fistula remained plump after arm elevation. The physical examination findings were categorized into four predictions of lesion location: inflow, outflow, both, or neither. Both the interventional specialist and student were masked as to the results of the other. Potential confounding variables, including age, gender, diabetic status, and location of AVF were recorded. The prediction was compared to the gold standard of AVF angiography. An independent third party reviewed the short video run of the angiograph and determined stenosis type.
Outflow
Inflow
Both
Neither
1.000 0.000 0.500 0.000
0.000 1.000 0.500 0.000
0.500 0.500 1.000 0.000
0.000 0.000 0.000 1.000
Forty-nine patients were examined in 6 months of patients referred with suspicion of fistula dysfunction. The average age of the subjects was 59.1 and majority of the subjects were male (61.2%), diabetic (53.1%), and had an upper arm AVF (75.5%). The full-time interventional specialist demonstrated correct prediction of lesion location of 89.8% (kappa = 0.850), while the undergraduate student had a correct prediction of 77.6% (kappa = 0.625). The student’s performance, however, differed significantly over time. The student correctly predicted the location of the lesion in 6 (42.9%) of the first 14 patients (kappa = 0.082), compared to 32 (91.4%) of the last 35 patients (kappa = 0.855). The interventional nephrologist correctly predicted the location of the lesion in 11 (78.6%) of the first 14 patients (kappa = 0.661), compared to 33 (94.3%) of the last 35 patients (kappa = 0.876).
546
Rosenberg et al.
TABLE 2. Weighted kappa values for physical examination by the undergraduate student and interventional nephrologist with expertise in physical examination Party Interventional specialist Undergraduate student
Correct diagnosis
Correct diagnosis in first 14 patients
Correct diagnosis in last 35 patients
91.9% kappa = 0.850 80.2% kappa = 0.625
78.6% kappa = 0.661 42.9% kappa = 0.082
94.3% kappa = 0.888 91.4% kappa = 0.830
There was substantial agreement beyond chance between the physical examination and the angiography in the diagnosis of outflow stenosis (85.7%, kappa = 0.697) by the undergraduate student and a moderate agreement beyond chance for prediction of inflow stenosis (79.6%, kappa = 0.583). The interventional nephrologist also had a substantial agreement beyond chance between the physical examination and the angiography in the diagnosis of outflow stenosis (93.9%, kappa = 0.863) and substantial agreement beyond chance in the diagnosis of inflow stenosis (81.6%, kappa = 0.874). When comparing lower arm fistulae in just the last 35 patients, the undergraduate student correctly predicted lesion location with a strong agreement beyond chance of 80.0% (kappa = 0.688), while the interventional specialist performed with a perfect agreement of 100.0% (kappa = 1.000). In upper arm fistulae, the student performed the exam with a higher level of accuracy than compared with the lower arm fistulae (93.33%, kappa = 0.854) and was in perfect agreement with the interventional nephrologist (93.33%, kappa = 0.854). Discussion This study demonstrates that an undergraduate student with no previous medical knowledge or physical examination experience can accurately predict lesion location in dysfunctional AVF at a comparable rate to that of a full-time interventional specialist (Figs. 1 and 2, Table 2). After examining 14 patients, the undergraduate’s ability to correctly diagnose lesion location was 91.4% for the subsequent 35 patients. The specialist’s ability to correctly predict fistula dysfunction by physical exam increased over time as well, suggesting performing exams will thus increase even a specialist’s ability to correctly predict fistula dysfunction. Correct prediction of outflow lesions was greater than that of inflow lesions. The reason for this is unclear, but may be related to the presence of other clinical features that provide diagnostic clues, including edema of the arm and shoulder, breast, supraclavicular region, neck and face, and presence of collateral vasculature on the chest wall. The results obtained support our hypothesis that the physical exam skill can be taught to a nonmedical professional in a relatively short amount of time. Although this study only examined location of stenosis in patients referred for intervention, detection of
stenosis in an unselected group may not be as easy but this information still proves useful. It follows that the exam of the vascular access can be taught to general nephrologists, dialysis technicians, and nurses that work in the hemodialysis unit. Teaching patients to assess their own fistula as well would also prove beneficial to increase access longevity by promptly reporting abnormal findings (15). While advanced technology such as ultrasound and static and dynamic venous pressure is often used, the physical exam is free and readily available and can be performed in less than 1 minute. Nephrology training centers should make the physical examination of vascular accesses an essential part of their training curriculum, as it would be extremely cost and time efficient. Future research should focus on educational interventions in the hemodialysis unit that promote appropriate and timely referral for angiography.
References 1. Leon C, Asif A: Physical examination of arteriovenous fistulae by a renal fellow: does it compare favorably to an experienced interventionalist? Semin Dial 21:557–560, 2008 2. Mishler R, Shon D, Hubert B, Nissenson AR: Development and usefulness of a physical examination toll to diagnose vascular access dysfunction. J Am Soc Nephrol 11:190A, 2000 3. Choi JR, Kim YS, Yoon SA, Won YD, Son YS, Song WJ, Song HC, Kim YS, Chang YS, Bang BK, Kim YO: Accuracy of physical examination in the detection of arteriovenous fistula dysfunction. Korean J Nephrol 25:797–802, 2006 4. Asif A, Cherla G, Merrill D, Cipleu CD, Briones P, Pennel P: Conversion of tunneled hemodialysis catheter-cosigned patients to arteriovenous fistula. Kidney Int 67:2399–2406, 2005 5. Asif A, Leon C, Orozco-Vargas LC, Krishnamurty G, Choi KL, Mercado C, Merrill D, Thomas I, Salman L, Artikov S, Bourgoignie JJ: Accuracy of physical examination in the detection of arteriovenous fistula stenosis. Clin J Am Soc Nephrol 2:1191–1194, 2007 6. Asif A, Leon C, Orozco-Vargas LC, Krishnamurty G, Choi KL, Mercado C, Merrill D, Thomas I, Salman L, Artikov S, Bourgoignie JJ: Accuracy of physical examination in the detection of arteriovenous graft stenosis. Semin Dial 21:85–88, 2008 7. National Kidney Foundation: K/DOQI clinical practice guidelines in vascular access: 2006 update. Am J Kidney Dis 48(Suppl. 1): S176–S306, 2006 8. Beathard GA: Physical examination of the dialysis vascular access. Semin Dial 11:231–236, 1998 9. Beathard GA: Physical examination: the forgotten tool. In: Gray R, Sands J (eds). A Multidisciplinary Approach for Hemodialysis Access. New York: Lippincott, Williams & Wilkins, 2002:11–118 10. Beathard GA: An algorithm for the physical examination of early fistua failure. Semin Dial 18:331–335, 2005 11. Beathard GA: Physical examination of the dialysis vascular access. Semin Dial 11:231–236, 1998 12. Beathard GA: A Practitioners Guide To Physical Examination of Dialysis Vascular Access. Dallas: The End Stage Renal Disease Network of Texas, Inc. 2003 13. McGinn T, Wyer PC, Newman TB, Keitz S, Leipzig R, Guyatt G: Tips for learns of evidence-based medicine: measures of observer variability (kappa statistic). CMAJ 171:1369–1373, 2004
PHYSICAL EXAMINATION OF AVF 14. Trerotola SO, Ponce P, Stavropoulos SW, Clark TWI, Tuite CM, Mondschein JI, Shlansky-Goldberg R, Freiman DB, Patel AA, Soulen MC, Cohen R, Alan Wasserstein A, Chittams JL: Physical examination versus normalized pressure ratio for prediction outcomes of hemodialysis access interventions. J Vasc Interv Radiol 14:1387–1393, 2003
547
15. Vassalotti JA, Jennings WC, Beathard GA, Neumann M, Caponi S, Fox CH, Spergel LM and the Fistula First Breakthrough Initiative Community Education Committee: Targeting Catheter Last in Fistula First. Semin Dial 25(3):303–310, 2012
