The refluxing anterior accessory saphenous vein demonstrates similar clinical severity when compared to the refluxing great saphenous vein

Abstract

Objective

To prospectively compare disease severity in subjects with anterior accessory saphenous vein versus great saphenous vein incompetence with an incompetent saphenofemoral junction.

Methods

Data were^ collected from 241 subjects and 290 limbs over a six-month period. These subjects were categorized into three groups with primary venous reflux disease, namely anterior accessory saphenous vein, great saphenous vein, and control. Statistical methods including descriptive statistics, student t-tests, and log linear modeling were employed to compare groups and determine predictive features out of the 41 demographic and disease-specific variables collected.

Results

Subjects in the anterior accessory saphenous vein group and those in the great saphenous vein group demonstrate statistically significant differences as compared to the control group with respect to the following disease-specific features: mean VCSS, presence of C2 and C3 disease. The anterior accessory saphenous vein group also showed statistically significant differences in gender compared to both great saphenous vein and control, as well as mean body mass index compared to the control. Log-linear modeling revealed equivalent disease severity when comparing patients with saphenofemoral junction reflux to the great saphenous vein or anterior accessory saphenous vein.

Conclusions

Patterns of reflux from the saphenofemoral junction to either the anterior accessory saphenous vein or great saphenous vein possess similar disease severity and commonly suffer complications of venous stasis.

Keywords

Classification of venous disease epidemiology great saphenous vein quality of life assessment varicose veins

Introduction

Venous hemodynamics and patterns of pathology are heterogeneous in nature with a variety that can be truly perplexing. The anterior accessory great saphenous vein (AASV) is a common pathway for both primary and recurrent symptomatic varices, which is often inadequately recognized by those unfamiliar with venous anatomy.^1–3 The purpose of this study was to test whether patients with saphenofemoral junction (SFJ) reflux to the great saphenous vein (GSV) share similar disease severity as those with reflux from the SFJ to the AASV.

Anatomically, the AASV vessel rests within the same layer of the limb as the GSV, lying just superficial to the deep venous system over the proximal anterior thigh. It has been shown that both the refluxing GSV and AASV patterns adversely affect patient-perceived quality of life measures.^4,5 Despite the common finding of AASV reflux, the majority of governing bodies focus on few vein patterns, namely the great saphenous and small saphenous veins.⁶ Ironically, the refluxing small saphenous vein occurs with approximately the same frequency as the AASV.

This paper prospectively compares disease severity through validated severity scoring tools in patients with primary SFJ reflux, the GSV or AASV versus control.

Methods

Patients

This prospective observational study registered consecutive patients at a dedicated vein center managing the spectra of venous and lymphatic pathology from November 2013 through February 2014. A single credentialed vascular sonographer in an accredited venous testing center performed all duplex assessments. Inclusion criteria required subjects to have primary venous reflux disease, a normal deep venous system, and no duplex signs of proximal obstruction, e.g. normal phasicity and mean velocities as measured at the common femoral vein. Consecutive subjects meeting criteria for inclusion were categorized to one of three groups. The patients in the AASV group had a reflux pattern from SFJ to anterior accessory saphenous vein and tributaries. The GSV group included only those with reflux from SFJ to GSV above the knee. The control group was defined by the presence of primary superficial venous reflux disease, absence of deep vein incompetence, not meeting the duplex reflux criteria, e.g. SFJ to the GSV or AASV. The control group possessed limbs with GSV incompetence and normal SFJ, isolated tributary disease with normal saphenous trunks, and those with burden of isolated reticular veins and telangiectasia. Patients with recurrent varices after intervention, small saphenous incompetence, and combined AASV/GSV reflux were excluded from consideration. Forty-one variables including demographics, duplex parameters, and disease severity measures were captured for each subject (Tables 1 and 2).

Table 1.

Data fields and formatting.

Field name	Data type	Valid values
Unique patient ID	VARCHAR	Character strings from first name, last name, gender, and birth year
DOS	DATE	Date (mm,dd,yyyy)
Age	INTEGER	Values greater than zero
Gender	BIT	Male = 0; female = 1
BMI	TEXT	Numeric values to single decimal point auto calculated in EMR
Primary disease	BIT	No prior treatment = 0; prior vein treatment = 1
Revised CEAP
C0	BIT	Feature not present = 0; Feature present = 1
C1	BIT	Feature not present = 0; Feature present = 1
C2	BIT	Feature not present = 0; Feature present = 1
C3	BIT	Feature not present = 0; Feature present = 1
C4a	BIT	Feature not present = 0; Feature present = 1
C4b	BIT	Feature not present = 0; Feature present = 1
C5	BIT	Feature not present = 0; Feature present = 1
C6	BIT	Feature not present = 0; Feature present = 1
VCSS	INTEGER	Values greater than zero and less than 31
Duplex physiology (anatomic CEAP-modified) Table 2	INTEGER	Reflux = 1; obstruction = 2; reflux and obstruction = 3; Normal = 4; successful ablation = 5
Vein diameters	NUMERIC	Values greater than zero to single decimal point (unit = mm)

DOS: date of service; BMI: body mass index; CEAP: clinical, etiologic, anatomic, and physiologic (CEAP) classification.

Table 2.

Duplex anatomy fields aligned with CEAP.

CEAP (anatomic value)	Region	Field name (separate columns for each lower extremity)
1	Superficial	Reticular veins and telangiectasia
2j	Superficial	Saphenofemoral junction
2a	Superficial	Anterior accessory saphenous vein
2p	Superficial	Posterior accessory saphenous vein
2	Superficial	Great saphenous vein above the knee
3	Superficial	Great saphenous vein below the knee
4j	Superficial	Saphenopopliteal junction
4	Superficial	Small saphenous vein
5a	Superficial	Varices above the knee
5b	Superficial	Varices below the knee
9	Deep	External iliac vein
11	Deep	Common femoral vein
12	Deep	Femoral profunda vein
13	Deep	Femoral vein
14	Deep	Popliteal vein
15	Deep	Crural veins (anterior tibial, posterior tibial, peroneal)
16	Deep	Muscular (gastroc and soleal)
17	Perforators	Incompetent perforators AK
18	Perforators	Incompetent perforators BK

CEAP: clinical, etiologic, anatomic, and physiologic (CEAP) classification.

Data capture and mining

Data capture was facilitated through MacPractice (MacPractice, Inc., Lincoln NE), an Apple Mac-based, electronic medical record (EMR) system. The EMR templates were specifically formatted for both ease of use in documentation and data capture, formatted to match fields in a vein specific registry.⁷ The backbone of this system is a MYSQL database (Oracle Corporation, Redwood City, CA). Raw data were mined to a comma-separated file using Sequel Pro for Mac (Software Design, Australia) and a defined script to pull targeted fields for analysis.

Data analysis

Descriptive statistics were utilized to assess all demographic and disease-specific variables independently for each group. The mean and standard deviation were calculated for the following variables: age, body mass index (BMI), VCSS, and vein diameter. In order to determine statistically significant mean differences between the three groups, student t-tests were performed for each of these variables. Gender, laterality, and disease-specific variables including C0, C1, C2, C3, C4a, C4b, C5, and C6 were all measured categorically, e.g. male/female or present/not present. For this reason, these variables were represented using frequency tables containing count information for each reflux pattern; significant associations and relationships were tested. All differences and relationships tested were identified as significant when the p-values were less than .05.

The aforementioned analyses assess significant differences between the three groups for each variable separately. In order to better understand the relationships and patterns of these variables and their impact on the groups, it is important to simultaneously assess the combinations and interactions of the variables together. Conditional independence models were constructed for this. Once completed, we were able to visualize a pattern. If the interaction patterns are similar, we may conclude that AASV and GSV are similar. Log-linear modeling is a standard approach applied to determine direction and magnitude of the relationship between more than two variables at a time.^8,9 This process simultaneously tests multiple contingency tables to express complex probabilistic relationships and odds ratios. In an iterative stepwise manner, unnecessary variables are removed from the model (Akaike criterion (AIC) was employed for this) leaving only the most important factors in the final model.

Results

Consecutive subjects from October 2013 through February 2014 were considered for inclusion. The patient sample included 241 patients and 290 limbs. There were 23 subjects in the AASV group, 145 in the GSV group, and 78 in the control group. Forty-four limbs failed to meet the criteria and were excluded from analysis. Table 3 highlights the demographic details for the three groups. Of significance, the AASV group comprised 86.9% females which was significantly different from to the GSV and control groups (71.8% and 65.0%, respectively, p < .05). The AASV group also shared a significantly greater BMI when compared to the GSV group and control (33.1 vs. 29.2 and 29.9, respectively, p < .05).

Table 3.

Demographic variables and comparisons.

Demographics	AASV	GSV	Control
N	23 (9.5%)	145 (60.2%)	78 (32.3%)
Gender (M:F)	3:20	43:102	17:61	Group 1 vs. Group 2, p < .05
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < .05
Laterality (R:L)	9:14	76:69	36:42	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < NS
Mean age (yr)	58.1	59.3	58.0	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < NS
Mean body mass index	33.1	29.2	29.9	Group 1 vs. Group 2, p < .05
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < .05

GSV: great saphenous vein; AASV: anterior accessory great saphenous vein; NS: not significant.

A revised clinical, etiologic, anatomic, and physiologic (CEAP) classification shown in Table 4 revealed significantly higher presence of C2 and C3 disease for the AASV and GSV groups when compared to the control group, p < .05, but not when compared to each other. Comparatively, VCSS was found to be significantly higher for AASV group subjects compared to the GSV and control groups, p < .05 and p < .00001, respectively. The GSV group also demonstrated significantly higher venous clinical severity score when compared to the control group, p < .05.

Table 4.

Disease-specific features and comparisons.

Disease-specific features	AASV	GSV	Control
Mean VCSS	8.0	6.0	3.9	Group 1 vs. Group 2, p < .05
				Group 2 vs. Group 3, p < .00001
				Group 1 vs. Group 3, p < .00001
C0	0 (0%)	2 (1.4%)	0 (0%)	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < NS
C1	4 (17.4%)	20 (13.8%)	24 (30.7%)	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < NS
C2	21 (91.3%)	125 (86.2%)	31 (39.7%)	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < .05
				Group 1 vs. Group 3, p < .05
C3	7 (30.4%)	35 (24.1%)	7 (9.0%)	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < .05
				Group 1 vs. Group 3, p < .05
C4a	2 (8.7%)	3 (2.1%)	1 (1.3%)	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < NS
C4b	1 (4.3%)	5 (3.4%)	0 (0%)	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < NS
C5	1 (4.3%)	5 (3.4%)	1 (1.3%)	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < NS
C6	1 (4.3%)	3 (2.1%)	1 (1.3%)	Group 1 vs. Group 2, p < NS
				Group 2 vs. Group 3, p < NS
				Group 1 vs. Group 3, p < NS
Mean vein diameter/
median (mm)	6.4/ (6.4)	7.2/ (6.8)	N/A	Group 1 vs. Group 2 p < NS
Superficial thrombophlebitis	3 (13.0%)	10 (6.8%)	2 (2.6%)

GSV: great saphenous vein; AASV: anterior accessory great saphenous vein; NS: not significant; VCSS: venous clinical severity score.

The presence of superficial thrombophlebitis was captured from duplex reporting. AASV group and GSV group had a prevalence of thrombotic complications approximating 10%, which equates to about 10 patients, in this sample of subjects with primary reflux disease. Because such a small sample of patients showed the presence of superficial thrombophlebitis, no significance testing was performed, and only descriptive statistics are reported in Table 4.

When analyzing the demographic and disease-specific variables simultaneously with log linear modeling, we find that AASV and GSV have statistically the same pattern of severity in signs and magnitude for the explanatory variables studied as shared in Table 5. Both AASV and GSV showed a significant prevalence for female gender and positive frequency as age and BMI increase. The AASV pattern is found at a significantly higher proportion in female subjects. What is also interesting is that the disease-specific variable C1 was found to be significantly less frequent in both AASV and GSV, while C3 is significantly more frequent for both groups. These similarities indicate that patients suffering from an incompetent SFJ to the AASV or GSV possess the same behavior.

Table 5.

Log-linear model results.

Variable	Estimate	p
Main effects
(Intercept)	−8.14	<.05
AASV	−16.75	NS
GSV	−16.36	NS
Female	1.24	<.05
Age 30–40	1.48	<.05
Age 40–50	1.86	<.05
Age 50–60	2.17	<.05
Age 60–70	2.10	<.05
Age 70–80	1.69	<.05
Age 80+	1.10	<.05
Normal BMI	1.79	<.05
Overweight BMI	1.87	<.05
Obese BMI	1.87	<.05
C0	−18.14	NS
C1	0.73	<.05
C2	−0.20	NS
C3	−3.66	<.05
Interaction effects
Interaction variable	AASV (Estimate/ significance)	GSV (Estimate/ significance)
C0	0.82 (NS)	14.02 (NS)
C1	−2.18 (<.05)	−2.45 (<.05)
C2	−2.45 (<.05)	1.92 (<.05)
C3	2.75 (<.05)	2.55 (<.05)

GSV: great saphenous vein; AASV: anterior accessory great saphenous vein; BMI: body mass index; NS: not significant.

Discussion

The incidence of AASV reflux in our sample was in line with other studies reported.^1–3 This suggests that our study population, though relatively small, was consistent with the general population, minimizing concern for single-center bias. This study is the first to compare clinical severity scores for the primary disease groups studied, demonstrating that a common and meaningful relation in severity exists for subjects with AASV or GSV incompetence. VCSS significance when compared to GSV incompetence might easily be explained by the burden of varices above and below the knee in patients with AASV incompetence. These patients often have complex reflux patterns explaining the significant near two-point disparity in severity scoring.

Females were predominately affected with AASV reflux in our sample. Women with central obesity retain sex hormones and secrete low doses of relaxin over time, contributing significantly to the etiology of venous incompetence.¹⁰ The obese limb also loses structural support for the superficial venous system, as fatty tissue infiltrates the saphenous sheath and subcutaneous tissues. These among other issues may explain truncal reflux but not necessarily that isolated to the AASV. Although not assessed in our study, additional influences of multiparity, occupational status, and others variables may yield more insight to the triggers of developing primary reflux in an AASV versus GSV.

Although we only studied those with primary disease, the AASV as a source of recurrence raises a number of questions no matter what treatment was employed. Does this vessel become problematic due to disease progression, a competing change in hemodynamics, or are we causing the AASV to become problematic? There are arguments for each of these positions, suggesting further study is needed to truly investigate this process.

Anatomically speaking, treating the AASV with thermal ablation is quite successful assuming that adequate energy (j/cm²) is delivered to the target vein.¹¹ The AASV has far fewer treatment influences when compared to the GSV, as there are few perforators with direct communication with this vessel beyond the SFJ.

The limitations of this manuscript include the relatively small sample size and the lack of disease specific and generic patient reporting measures.¹² In an attempt to assess a true burden of disease, we cannot underestimate the value of the patient’s perspective both before and after treatment. Questions in medicine may be answered more quickly in daily practice through practical clinical data collection without the cost and work involved through implementing a randomized trial.¹³ This study is an example where formatted data routinely captured in real world day-to-day activities possesses powerful information.¹⁴ It is inevitable that future epidemiology and outcome studies will require active subject participation, as they share the true impact of a disease process or treatment received.¹⁵

Conclusions

Primary incompetence of anterior accessory veins is common with disease severity and morbidity that matches that of the incompetent GSV. Patients with isolated reflux of the AASV should have the same opportunity for treatment as those with GSV incompetence.

Footnotes

Acknowledgements

The authors wish to recognize the valuable statistical support and professional medical editing by Meghan Honerlaw Tooman.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Labropoulos

Kang

Mansour

. Primary superficial reflux with competent saphenous trunk. Eur J Vasc Endovasc Surg 1999; 18: 201–206.

Garcia-Gimeno

Rodriguez-Camarero

Tagarro-Villalba

. Duplex mapping of 2036 primary varicose veins. J Vasc Surg 2009; 49: 681–689.

Rasmussen

Bjoern

Lawaetz

. Randomised clinical trial comparing endovenous laser ablation with stripping of the great saphenous vein: clinical outcome and recurrence after 2 years. Eur J Vasc Endovasc Surg 2010; 39: 630–635.

Bush RG, Flanagan J, Gueldner T, et al. Factors associated with recurrence of varicose veins after thermal ablation: results of the recurrent veins after thermal ablation study. Sci World J 2014; Article ID 505843: 7.

Theivacumar

Darwood

Gough

. Endovenous laser ablation (EVLA) of the anterior accessory great saphenous vein (AAGSV): abolition of sapheno-femoral reflux with preservation of the great saphenous vein. Eur J Vasc Endovasc Surg 2009; 37: 477–481.

Treatment of Varicose Veins. Anthem blue cross blue shield web site, www.anthem.com/medicalpolicies/policies/mp_pw_a053323.htm (accessed 19 May 2014).

Chi

Schul

Gibson

. Chronic venous disorder registry: a new perspective. Phlebology 2014; 29: 415027–415027.

Agresti

. Log linear model for contingency tables. An introduction to categorical data analysis, 1st ed. Hoboken, NJ: Wiley & Sons, 2007, pp. 204–222.

Faraway

. Chapter 4. Extending the linear model with R, Boca Raton, FL: Chapman & Hall/CRC, 2005, pp. 89–97.

10.

Beebe-Dimmer

Pfeifer

Engle

. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidem 2005; 15: 175–184.

11.

Proebstle

Krummenauer

Gül

. Nonocclusion and early reopening of the great saphenous vein after endovenous laser treatment is fluence dependent. Dermatol Surg 2004; 30: 174–178.

12.

Michaels

Brazier

Campbell

. Randomized clinical trial comparing surgery to conservative treatment for uncomplicated varicose veins. Br J Surg 2006; 93: 175–181.

13.

Horn

Gassaway

. Practice Based evidence: incorporating clinical heterogeneity and patient-reported outcomes for comparative effectiveness research. Med Care 2010; 48: S17–S22.

14.

Tunis

Stryer

Clancy

. Practical clinical trials: increasing the value of clinical research for decision making in clinical and health policy. JAMA 2003; 290: 1624–1632.

15.

Couto

Comer

. Patient engagement: the critical catalyst to health reform in the USA. J Comp Eff Res 2012; 1: 209–211.