845

Effect of Anatomic
Variations on Deep Venous
Thrombosis of the Lower
Downloaded from www.ajronline.org by 36.81.54.252 on 10/18/20 from IP address 36.81.54.252. Copyright ARRS. For personal use only; all rights reserved

Extremity

I , ,, . .

Gin-Chung Liu1 Three hundred thirty-seven lower extremities of 256 patients, both symptomatic and
Ernest J. Ferris asymptomatic, underwent ascending positive contrast venography. Certain anatomic
John R. Reifsteck variations were noted to be related to the location and development of deep venous
thrombosis (DVT). The thrombus was located entirely in the calf in 44% of asymptomatic
Max E. Baker
limbs and 17% of symptomatic ones (p < 0.001). There were multiple superficial femoral
veins (SFVs) in 31% of the limbs examined, and 40% of those
had DVT. This limbs
percentage (40%) represented a statistically higher incidence
< 0.001) than of DVT (p
those limbs with a single SFV, where the incidence was only 19%. When limbs with
multiple SFVs had DVT, only 41% were symptomatic, whereas in those with a single
SFV, 72% of the limbs were symptomatic (p < 0.001). This difference may be the result
of internal collaterals due to multiple SFVs. When there were more than five valves in
the deep veins between the poplfteal fossa and the ischial spine, there was a higher
incidence of DVT (35%) as compared to the incidence when there were five or fewer
valves (13%) (p < 0.001).

In spite of the advancement of various noninvasive methods, conventional

positive contrast ascending venography remains the definitive diagnostic exami-
nation in patients with suspected deep venous thrombosis (DVT) [1 -3]. The
purpose of this study was to determine whether there were any anatomic variations
that might predispose to DVT and to explain why so many patients with DVT are
asymptomatic [4-6]. Our experience includes four different categories of patients:
(1) clinically symptomatic but venographically negative, (2) clinically symptomatic
and venographically positive, (3) clinically asymptomatic but venographically posi-
tive, and (4) clinically asymptomatic and venographically negative for DVT.

Materials and Methods

Conventional ascending positive contrast venography of lower extremities was performed

in 337 limbs of 256 patients, using the technique recommended by Rabinov and Paulin [71.
In conjunction with the Department of Orthopedic Surgery, we have developed an investiga-
tional protocol of performing preoperative and 5-7 day postoperative bilateral leg venograms
on all patients with hip fractures or total hip replacements because of the high incidence of
Received September 3. 1985; accepted after DVT reported in these patients [5, 6]. Hence, a large series of asymptomatic patients was
revision December 26, 1985. included. We have, of course, also evaluated many symptomatic patients with clinical
Presented at the annual meeting of the American symptoms and signs of DVT. Thus, we have been able to divide our patients into four
Roentgen Ray Society. Washington, DC. April categories as mentioned above.
1986.
Variations of the superficial femoral vein (SFV) and popliteal vein (PV) anatomy were
1 All authors: Department of Radiology, Univer- observed. The number of the valves of the deep veins from the level of the popliteal fossa up
sity of Arkansas for Medical Sciences, 4301 W.
to the ischial spine region were also tabulated. Statistical analysis was obtained by using the
Marhkam, Little Rock, AR 72205. Address reprint
requests to E. J. Ferris. chi-square test with Yates’ correction [81 to determine the significance of variations of vein
numbers (fig. 1) and differences in the number of valves in the SFV. The correlations between
AJR 146:845-848, April 1986
0361 -803X186/1464-0845 different categories as deScribed above and incidence of DVT were analyzed with respect to
C American Roentgen Ray Society these two factors.
 846 LIU ET AL. AJR:146, April 1986

TABLE 1 : Distribution of Thrombi by Anatomic Site

Thrombotic Site Asymptomatic (%) Symptomatic (%)

Pelvic veins only 0 1 (1)

Thigh and pelvic veins 0 4 (5)
Thigh veins only i 4 (39) 25 (31)
Thigh and calf veins 3 (8) 37 (46)
i
Downloaded from www.ajronline.org by 36.81.54.252 on 10/18/20 from IP address 36.81.54.252. Copyright ARRS. For personal use only; all rights reserved

Calf veins only i 9 (53)” 1 4 (17)”

Total 36 (i 00) 81 (100)
“p<O.O0.

DVT(+)
DVT(-)
Total 124
25
99
2 TCthD
TABLE
and Multiple
2: Relation of Deep Venous Thrombosis
Superficial Femoral Veins

Single
sFv Number
between

Multiple Total
Single

DVT (+):
Asymptomatic (A) 11 22 33
Symptomatic (B) 28 15 43
Subtotal (C) 39 37 !__

DVT (-):
Asymptomatic (D) 123 34 157
Symptomatic (E) 4i 22 63
Subtotal
Note.-DVT = (F)
Deep venous thrombosis; SFV =1 64
superficial 56
femoral vein. A:D. 220
A:B. C:F,
a

H IE II\ iF IIIIG I\ IH p<O.001.

DVT(+) 7 14 1 0 TABLE 3: Relation between Valve Number and Patient Groups

DVT(-) 14 63 3 2
Valve Number
Total 21 77 4 2
55 >5 Total
Fig. 1 -Anatomic variations of superficial femoral vein (SFV) and popliteal
DVT (+):
vein (PV) in 296 limbs (percentage of positive deep venous thrombosis) A,
Single SFV (20#{176}h).
B, Fenestration of SFV (35%). C, Multiple accessory SFV
Asymptomatic (A) 11 19 30
(60%). D, Single accessory SFV (33%). E, Duplication of 5FV and PV (33%). Symptomatic (B) 12 i6 28
F, Dupiication of v (18%). G, Multiple accessory SFV and PV (25%). H, Subtotal (C) 23 35 58
Multiple accessory PV (0).
DVT (-):
Asymptomatic (D) 1 10 47 157
Results Symptomatic (E) 46 i7 63
Subtotal (F) 1 56 64 220
Of 337 examinations performed, we found 296 satisfactory
limbs for evaluating anatomic variations, and 278 limbs de- Note.-c:F, p < O.0O.

tailed sufficiently to assess the number of valves present. Of

the clinically symptomatic limbs, 44% were free of thrombi on
venography. Of the clinically asymptomatic limbs, 19% had limbs with a single SFV were symptomatic (p < 0.001) (table

thrombi demonstrated on venography. 2). We found no difference relative to symptomatology and

incidence of DVT between single and multiple popliteal veins.
Table 1 analyzes the 1 1 7 instances of thrombi and shows
the distribution of DVT in the symptomatic and asymptomatic Table 3 shows that 60% of limbs with thrombi had more than
five valves, whereas only 29% of limbs without thrombi had
groups. In 82% of the symptomatic limbs, the thrombus was
located at least partly in the thigh. In the asymptomatic more than five valves (p < 0.001).
groups, only 47% of the limbs showed thrombi in the thigh,
with 53% having thrombi limited to the calf. Therefore, there Discussion
was a statistically significant difference in the location of
thrombi in the symptomatic vs. asymptomatic groups (p < The false-positive rate of clinical symptoms for DVT verified
0.001). by venography was 46%-67% in previous reports [9-1 1] and
For purposes of statistical analyses, both the SFV and PV was 44% in our study. One also should note that 1 9% of our
were divided into single and multiple groups. There were clinically asymptomatic limbs demonstrated thrombi on yen-
multiple SFVs in 31 % of the total limbs we examined. Of the ography. This finding is similar to that of a previous report of
limbs with multiple SFVs, 40% had DVT. This was a statisti- 22% [91. Our data again support the assertion that clinical
cally higher incidence (p < 0.001) than that seen in the 1 9% symptoms alone are not reliable for the diagnosis of DVT [7,
of those limbs with a single SFV. Only 41 % of the DVT limbs 9, 1 1 1. As previously reported, 29% of thrombi are located
with multiple SFVs were symptomatic, whereas 72% of DVT exclusively in the calf [9]. In our study, 28% of the total
 AJR:146, April 1986 DEEP VENOUS THROMBOSIS OF THE LEG 847

Fig. 2.-Thrombosis of superficial

femoral vein
(SFV). Leg venogram shows
filling defect in upper SFV (arrows) that
completely occludes lumen. Thrombus
apparently propagated from one branch
(large asterisk) of duplicated SFV with
sparing of other (small asterisk). This pa-
tient, probably because of this internal
Downloaded from www.ajronline.org by 36.81.54.252 on 10/18/20 from IP address 36.81.54.252. Copyright ARRS. For personal use only; all rights reserved

collateral, had no symptoms. Venography

was performed because of positive radio-
nuclide venogram, routine in our depart-
ment when lung perfusion scan is per-
formed for diagnosis of pulmonary em-
bolism.

Fig. 3.-Deep venous thrombosis in

two patients: both patients illustrate mul-
tiple superficial femoral veins. A, Throm-
bus (arrows) in one channel of asympto-
matic limb. B, Multiple channels are in-
volved (arrows) in painful swollen limb.

thrombi were also confined to the calf. However, our addi-

tional information demonstrates that the asymptomatic DVT
limbs have a statistically higher incidence (53%) of pure calf
thrombi than symptomatic ones (1 7%). A previous report
showed no specific clinical feature relating to proximal and
distal thrombi [9]. Our findings support Piulachs’ theory [12],
which asserts that the initial subfascial and later epifascial
edema distal to the thrombus causes the clinical symptoms
including pain and swelling. The limb with the thrombus
confined to the calf has much better collateral circulation, so
it is less prone to develop subfascial edema and hence may
not develop epifascial edema. Obviously, this failure to de-
velop edema results in “silent” DVT.
We found another factor responsible for absence of symp-
toms, namely, multiple superficial femoral veins. In these
patients, complete occlusion by the thrombus is less likely
because multiplicity may offer internal collaterals, preventing
epifascial edema (fig. 2). Our study, furthermore, showed that
multiple-SFV limbs had a higher chance of having DVT (40%)
(fig. 3) than single-SFV limbs (1 9%). This is in contrast to
those limbs with multiple popliteal veins, where no such
incidence occurred. Perhaps the high incidence of DVT in
multiple SFVs can be explained by increased blood volume in
the venous pool and conversely a decreased flow rate, which
probably predisposes the limb to DVT [1 3, 14].
Limbs with DVT also had a greater likelihood of having a
valve number greater than five (fig. 4) than normal limbs (60%
vs. 29%). This suggests that the more valves a limb has, the
more likely venous stasis occurs. This is supported by pre-
vious work, which showed that intermittent changes in veloc-
ity, instead of sustained increases, for the purpose of emp- vein.
 848 LIU ET AL. AJR:146, April 1986

tying the valve cusps is effective for preventing postoperative 4. Kakkar VV. Prevention of venous thromboembolism. Clin Hae-
DVT [1 5, 1 6]. That the limb with more valves has a higher matol 1981;iO:543-58i
risk of DVT provides the real evidence that stasis [4] is a 5. Barnes RW, Brand RA, Clarke W, Hartley N, Hoak JC. Efficacy
of graded-compression antiembolism stockings in patients
causative factor for clinical thrombosis as hypothesized by
undergoing total hip arthroplasty. Clin Orthop 1978;i32:6i-67
Virchow in 1854.
6. Borow M, Goldson HJ. Prevention of postoperative deep venous
In summary, because of the unreliability of clinical symptom-
thrombosis and pulmonary emboli with combined modalities. Am
Downloaded from www.ajronline.org by 36.81.54.252 on 10/18/20 from IP address 36.81.54.252. Copyright ARRS. For personal use only; all rights reserved

atology of DVT, venography is necessary before starting Surg 1983;49:599-605

treatment. 7. Rabinov K, Paulin S. Roentgen diagnosis of venous thrombosis
Our study demonstrates that limbs with multiple SFVs and in the leg. Arch Surg 1972;i 04: 134-i 44
more than five valves between the level of the popliteal fossa 8. Steel RGD, Tome JH. Principles and procedures of statistics.
and ischial spine are at much higher risk for thrombosis. New York: McGraw-Hill, 1960
Patients with multiple SFVs, however, have a lower incidence 9. O’Donnell TF, Abbott WM, Athanasoulis CA, Millan VG, Callow
of clinical symptoms most likely relative to internal collateral. AD. Diagnosis of deep venous thrombosis in the outpatient by
venography. Surg Gynecol Obstet 1980;i 50:69-74
Finally, the asymptomatic DVT limbs had a higher incidence
10. Simpson FG, Robinson PJ, Bark M, Losowsky MS. Prospective
of thrombi confined to the calf. This anatomic-pathologic
study of thrombophlebitis and “pseudothrombophlebitis.” Lancet
correlation study makes prediction of potential DVT and its 1980;i :331 -333
characteristics possible. One may wish to consider the advis- 11 Cranley JJ, Canos AJ, Sull WJ. The diagnosis of deep venous
.

ability of performing preoperative venography on clinically thrombosis. Arch Surg 1976;i 1 1 :34-36
high-risk patients. 12. Nissl A. Acute thrombosis. In: May R. Surgery of the veins of the
leg and pelvis. Stuttgart: Georg Thieme, 1979:96-97
i3. Sigel B, Edelstein AL, Felix R, Memhardt CR. Compression of
the deep venous system of the lower leg during inactive recum-
REFERENCES
bency. Arch Surg 1973;i06:38-43
1 Gallus
. AS. Established venous thrombosis in pulmonary embo- 14. Sevitt S. Etiology and pathogenesis of deep vein thrombosis.
lism. Clin 1981;i 0:583-611
Haematol ‘Lancet 1960;i :384
2. SandIer DA, Duncan JS, Ward P. Diagnosis of deep-vein throm- 15. Browse NL, Jackson BT, Mayo ME, Negus AD. The value of
bosis: comparison of clinical evaluation, ultrasound, plethysmog- mechanical methods of preventing postoperative calf vein throm-
raphy, and venoscan with x-ray venogram. Lancet 1984;2 :716- bosis. Br J Surg 1974;6i :219-223
719 16. Nicolaides AN, Fernandes FE, Pollock AV. Intermittent sequential
3. Ramchandani P, Soulen RL, Fedullo LM, Gaines VD. Deep vein pneumatic compression of the legs in the prevention of venous
thrombosis: significant limitations ofnoninvasive tests. Radiology stasis and postoperative deep venous thrombosis. Surgery
1985;i 56:47-49 1980;87:69-76
 This article has been cited by:

1. Kyu‐Ho Yi, Hee‐Jin Kim. 2020. Is variation in posterior tibial veins a risk factor for deep‐vein thrombosis?. Clinical Anatomy
13. . [Crossref]
2. Maxim E. Shaydakov, Jose A. Diaz, Anthony J. Comerota, Fedor Lurie. 2020. Targeted gene expression analysis of human deep
veins. Journal of Vascular Surgery: Venous and Lymphatic Disorders . [Crossref]
3. A. Karasu, A. Šrámek, F. R. Rosendaal, R. J. van der Geest, A. van Hylckama Vlieg. 2018. Aging of the venous valves as a new risk
Downloaded from www.ajronline.org by 36.81.54.252 on 10/18/20 from IP address 36.81.54.252. Copyright ARRS. For personal use only; all rights reserved

factor for venous thrombosis in the elderly: the BATAVIA study. Journal of Thrombosis and Haemostasis 16:1, 96-103. [Crossref]
4. Ronald J. Gordon, Frederick W. Lombard. 2017. Perioperative Venous Thromboembolism. Anesthesia & Analgesia 125:2, 403-412.
[Crossref]
5. Melissa Andreia de Moraes Silva, Hanna Fatima Paranaíba Mesquita, Iara Gabriel Carneiro, Arturo Eduardo Krupa, Seleno Glauber
de Jesus Silva, Rodolfo Souza Cardoso. 2016. Variação anatômica venosa rara em membros inferiores. Jornal Vascular Brasileiro
15:4, 334-338. [Crossref]
6. Santosh K. Sangari. Veins of the Lower Limb 900-909. [Crossref]
7. E. Victoria Dydek, Elliot L. Chaikof. 2016. Simulated thrombin responses in venous valves. Journal of Vascular Surgery: Venous
and Lymphatic Disorders 4:3, 329-335. [Crossref]
8. Norio Uchida. 2016. Popliteal Venous Aneurysm Associated with Duplicated Popliteal Veins: Report of a Case. The Japanese
Journal of Phlebology 27:3, 365-369. [Crossref]
9. Angel López-Candales. Pulmonary Embolism 151-200. [Crossref]
10. Nigel Mackman. 2012. New insights into the mechanisms of venous thrombosis. Journal of Clinical Investigation 122:7, 2331-2336.
[Crossref]
11. Edwin G. Bovill, Albert van der Vliet. 2011. Venous Valvular Stasis–Associated Hypoxia and Thrombosis: What Is the Link?.
Annual Review of Physiology 73:1, 527-545. [Crossref]
12. P Paraskevas. 2011. Femoral vein duplication: incidence and potential significance. Phlebology: The Journal of Venous Disease 26:2,
52-55. [Crossref]
13. W. E. Trotman, D. J. Taatjes, P. W. Callas, E. G. Bovill. 2011. The endothelial microenvironment in the venous valvular sinus:
thromboresistance trends and inter-individual variation. Histochemistry and Cell Biology 135:2, 141-152. [Crossref]
14. Eun-Ah Park, Jin Wook Chung, Whal Lee, Yong Hu Yin, Jongwon Ha, Sang Joon Kim, Jae Hyung Park. 2011. Three-
Dimensional Evaluation of the Anatomic Variations of the Femoral Vein and Popliteal Vein in Relation to the Accompanying
Artery by Using CT Venography. Korean Journal of Radiology 12:3, 327. [Crossref]
15. Ivan Benaduce Casella, Calógero Presti, Yumiko Yamazaki, Alecxander A Vassoler, Luiz A Furuya, Claudio D Sabbag. 2010. A
duplex scan-based morphologic study of the femoral vein: Incidence and patterns of duplication. Vascular Medicine 15:3, 197-203.
[Crossref]
16. Chris M. Lindquist, Fern Karlicki, Patrick Lawrence, Jacek Strzelczyk, Neal Pawlyshyn, Iain D. C. Kirkpatrick. 2010. Utility of
Balanced Steady-State Free Precession MR Venography in the Diagnosis of Lower Extremity Deep Venous Thrombosis. American
Journal of Roentgenology 194:5, 1357-1364. [Abstract] [Full Text] [PDF] [PDF Plus]
17. Maria José García-Fuster, María José Fabiá, María José Forner, Alberto Cuñat. 2009. Deep venous thrombosis in duplicated
superficial femoral veins. Thrombosis Research 124:3, 379-380. [Crossref]
18. Colin Patrick Cantwell, Andrea Cradock, John Bruzzi, Patricia Fitzpatrick, Stephen Eustace, John G. Murray. 2006. MR
Venography with True Fast Imaging with Steady-state Precession for Suspected Lowerlimb Deep Vein Thrombosis. Journal of
Vascular and Interventional Radiology 17:11, 1763-1770. [Crossref]
19. . Lower Extremity Veins 374-393. [Crossref]
20. Daniel J. Quinlan, Raza Alikhan, Philip Gishen, Paul S. Sidhu. 2003. Variations in Lower Limb Venous Anatomy: Implications
for US Diagnosis of Deep Vein Thrombosis. Radiology 228:2, 443-448. [Crossref]
21. Gilbert B. Cushner, Fred D. Cushner, Michael A. Cushner. Deep Vein Thrombosis (DVT) and Total Knee Arthroplasty 694-704.
[Crossref]
22. Stewart, Manson, Dennis, Allan, Ludlam. 2000. Thrombosis in a duplicated superficial femoral vein in a patient with haemophilia
A. Haemophilia 6:1, 47-49. [Crossref]
23. Michael B. Gotway, Keith J. Edinburgh, Vickie A. Feldstein, Jonathan D. Lehman, Gautham P. Reddy, W.Richard Webb. 1999.
Imaging evaluation of suspected pulmonary embolism. Current Problems in Diagnostic Radiology 28:5, 133-184. [Crossref]
 24. N.J. Screaton, J.H. Gillard, L.H. Berman. 1998. The radiological investigation of suspected lower limb deep vein thrombosis.
Clinical Radiology 53:5, 388. [Crossref]
Downloaded from www.ajronline.org by 36.81.54.252 on 10/18/20 from IP address 36.81.54.252. Copyright ARRS. For personal use only; all rights reserved