European Journal of Obstetrics & Gynecology and Reproductive Biology 144 (2009) 173–176

Contents lists available at ScienceDirect

European Journal of Obstetrics & Gynecology and

Reproductive Biology
journal homepage: www.elsevier.com/locate/ejogrb

Deep venous thrombosis in gynecological oncology: Incidence and clinical

symptoms study
Joseph T. Santoso a,*, Leslie Evans b, Lauren Lambrecht b, Jim Wan b
a
University of Tennessee - West Clinic, 1588 Union Avenue, Memphis, TN 38104, United States
b
University of Tennessee, United States

A R T I C L E I N F O A B S T R A C T

Article history: Objective: Cancer patients have increased risks of leg deep venous thrombosis (DVT). We studied the
Received 28 September 2008 incidence, risk factors and most predictive symptoms of leg DVT in gynecologic oncology patients.
Received in revised form 13 March 2009 Study designs: Gynecologic oncology patients with any leg DVT symptoms were recruited and screened
Accepted 20 March 2009
using Doppler sonogram. All hospitalized surgery and non-ambulating patients received thigh-high
sequential compression devices (SCDs) without heparin as a prophylactic method against thrombosis.
Keywords: Statistical analysis was done using chi-square or Fisher’s exact tests.
Deep venous thrombosis
Results: Out of 1974 patients, 134 complained of lower limb symptoms. Doppler studies found 38
Gynecology oncology
patients with leg DVT. Incidence of leg DVT was 36/853 (4.2%) in patients with cancer and 2/1121 (0.2%)
Cost
in patients without cancer (odds ratio 2.8 with a diagnosis of cancer). Leg edema, erythema, fever, and
warm leg were significant symptoms in diagnosing leg DVT (p < 0.01). The cost of finding a leg DVT was
$747.54.
Conclusions: Clinical exam is less accurate than Doppler sonogram in diagnosing leg DVT. The incidence
of leg DVT using SCD seems to be comparable with other studies. Finally, the cost of identifying leg DVT
seems reasonable.
ß 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction comparable with unfractionated heparin 5000 units TID heparin

Gynecologic oncology patients have high risks for developing
deep venous thrombosis (DVT) since they experience hypercoa-
gulable states, immobility and vascular injuries during the course
of their treatment. The risk of perioperative DVT was reported with
ranges from 19.6% to 38% in patients with gynecologic cancers
versus 10–15% in benign gynecology [1,2]. The best method of DVT
prophylaxis in this population is still controversial. A controlled
trial of two unfractionated heparin regimens for prevention of
postoperative DVT showed 5000 units heparin three times daily
decreased risk of DVT from 18.4% to 8.7% [3]. The ENOXACAN II
study found that 4 weeks versus 1 week of low molecular weight
heparin reduced the rate of the DVT from 12% to 4.8% [4].
Comparing unfractionated heparin 5000 units three times daily
versus pneumatic compression, Maxwell et al. found no difference
in incidence of DVT but a higher transfusion rate for patients
receiving heparin [5].
In our gynecologic oncology practice, we use a thigh-high
sequential compression device (SCD) for leg DVT prophylaxis
based on the Clark-Pearson study showing SCD (4% DVT rate) as
* Corresponding author. Tel.: +1 901 322 0251; fax: +1 901 322 0259.
E-mail address: jsantoso@westclinic.com (J.T. Santoso).
prophylaxis (6% DVT rate) [6]. However, the first ENOXACAN study
in postoperative abdominal surgery patients showed a much
higher DVT rate (16.5%) despite using 5000 units unfractionated
heparin three times a day prophylaxis for 1 week [7]. The follow-up
ENOXACAN II study recommended 1 month of low molecular
weight heparin for postoperative patients to lower DVT rate to 4.8%
[4]. These two studies prompted us to look at our practice closely
and question if we need to change our practice from sequential
compression devices to the recommended prolonged usage (1
month) of low molecular weight heparin for leg DVT prophylaxis.
In our study, we wished to identify the incidence of leg DVT and
to confirm if sequential compression device usage alone without
heparin may result in an excessive rate of leg DVT. We also wished
to calculate the cost of leg DVT diagnosis and to evaluate predictive
clinical symptoms for leg DVT.
2. Materials and methods
Our Gynecologic Oncology division is a private practice-
academic combination consisting of six gynecologic oncologists,
four rotating residents, and several medical students. However, the
study patients were only derived from one gynecologic oncologist
since all of his patients are registered in his own database. The data
from each patient, from June 2001 to June 2005, were entered

0301-2115/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.ejogrb.2009.03.012
174 J.T. Santoso et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 144 (2009) 173–176

prospectively in real-time fashion into our own database (DDH Table 1

Diagnoses of patients seen in study period.
Software, Inc., Wellington, FL) to include demographic data,
oncology history and surgery findings. Follow-up on these patients Cancer Benign conditions
ranged from 6 months to 4 years. Included in the database were
Breast 19 Cervical dysplasia 153
any patients with symptoms of leg DVT. On a daily basis, the Cervix 289 Vaginal dysplasia 31
database was checked for completeness and accuracy. The Ovary/fallopian tube 195 Vulvar dysplasia 104
Methodist University Institutional Review Board approved the Placenta 14 Anal dysplasia 2
Uterus 255 Intra-op consulta 33
study.
Vagina 6 Pelvic pain 15
Patients suspected for having leg DVT with the clinical signs or Vulva 36 Benign cervix 19
symptoms of leg edema, tenderness, erythema and fever were Other site(s)b 39 Benign ovary 308
worked up in similar fashion. Fever was defined as equal to or Benign uterus 173
above 38 8C on examination. A thorough history was taken, signs Benign vagina 9
Benign vulva 37
and symptoms were noted, and an exam was performed by the
Others 91
resident physician and confirmed by the attending physician. If leg No surgery/pathc 146
DVT was suspected, lower limb ultrasound was ordered for both Totals 853 1121
extremities. The technique consisted of real-time ultrasound a
Intraoperative consult: mostly for bleeding, adhesion, injuries to bowel, ureter,
compression technique with duplex and color Doppler imaging. bladder during benign gynecologic surgeries.
DVT diagnosis was made if incomplete compressibility of a vessel b
Other sites: colon, rectum, appendix, lung, skin, urethra, kidney.
c
was found or absence of flow was noted on pulsed or color Doppler. No Surgery/pathology available: patients returned to the referring physicians,
The ultrasound findings were recorded in the database. lost to follow-up, or had operations elsewhere.

Patients were excluded if they were already being antic-

oagulated (taking aspirin, warfarin, heparin). We also excluded
patients with upper extremity DVT and pulmonary embolism since
their very low negligible incidence, in our population, causes
meaningless insignificant analysis.
During the study period, all patients received uniform DVT
prophylaxis. Before induction of anesthesia, the patients were
fitted with TED hose and sequential compression devices, which
they wore before, during, and after surgery until mobility was
again possible. Early ambulation was encouraged and usually
occurred on postoperative day one. Patients admitted for non-
surgical reasons were also fitted with sequential compression
devices until they were ambulatory. No prophylactic heparin or
coumadin was given.
Statistical analysis was done using either chi-square or Fisher’s
exact tests when appropriate. P-value < 0.05 was considered
statistically significant. The cost of Doppler sonogram for both legs
was calculated using CPT code 93670 with Medicare 2005
reimbursement of $211.99.
3. Results
During the 4-year study period, 1974 new patients to the
practice, all women, were seen. Out of this total, 853 patients were
diagnosed with cancer and 1121 patients were seen for benign
diagnoses (Table 1). The average age of the patients was 50.6 years
old with a range of 13–101 years old. There were seven Asian, 935
black, 20 Hispanic, 996 white patients and 16 other races in the
study population. We identified four pulmonary emboli and five
upper extremities DVT (mostly related to Hemaport). None of the
four patients with pulmonary embolism were diagnosed with DVT.
These nine patients were excluded from our analysis.
One hundred thirty-four out of 1974 patients were suspected to
have leg DVT because of symptomatic lower limb complaints;
therefore, they received leg Doppler ultrasound studies (Table 2).
The overall incidence of leg DVT was 4.2% (36/853 patients) in
patients with gynecologic malignancies and 0.2% (2/112 patients)
in patients with benign diagnoses. Only six of the patients with a
gynecological malignancy developed leg DVT within 4 weeks of
surgery (two after tandem and ovoid, three after total abdominal
hysterectomy/bilateral salpingoophorectomy/lymphadenectomy,
and one after palliative colostomy).
For the 853 patients with cancer, the chi-square p-value is
0.6285 when we compared the leg DVT rate between this group
with and without surgery. For the 1121 patients without cancer,
the Fisher’s exact p-value is 0.5401 when we compared the leg DVT
rate between this group with and without surgery. Therefore,
cancer diagnosis is a much more important risk factor than surgery
in developing lower limb DVT. The odds ratio of developing a leg
DVT with a diagnosis of cancer was 2.82 (p-value of 0.231) when
compared with benign diagnoses. In regards to specific cancer
diagnosis, patients with cervical cancer had a trend of developing
more leg DVT, but this was not statistically significant (Table 4).
Using the Fisher’s exact test, the p-value for this table is 0.9207.
In our study population, 134 leg Doppler ultrasounds were
ordered to find 38 legs DVT (28.4%). In other words, approximately
four clinical suspicions led to one confirmed leg DVT diagnosis. The
2005 Medicare rate for Doppler ultrasound (CPT code 93670) was
$211.99. Thus, the total cost of ordering 134 sonograms was
$28,406.66. The cost of finding each lower limb DVT was $747.54
($28,406.66 / 38 DVT).
On clinical exam, leg warmness, edema, fever and erythema of
the leg were important signs in diagnosing leg DVT (Table 3). In our
study, unilaterality and leg pains were not a strong predictive in
diagnosing leg DVT.
4. Discussion
One of the reasons that we embarked on this study was to
evaluate if we need to change our practice from using sequential
compression devices to 1 month of low molecular weight heparin
as leg DVT prevention. We feel that the DVT rate of 4.2% in our
cancer patients is comparable with many other studies [4,7–9].

Table 2
Patients in study period stratified by cancer, surgical, and DVT status.

853 patients with cancer 1121 patients without cancer

684 surgical 169 nonsurgical 720 surgical 401 nonsurgical

Patients with DVT symptoms 101 18 9 6

Doppler confirmed DVT 30 6 2 0
Incidence of leg DVT (%) 4.40% 3.60% 0.30% 0%
 J.T. Santoso et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 144 (2009) 173–176 175

Table 3
Clinical symptoms and their correlation to presence of leg DVT.

Sign/symptom Thrombus (N = 38) No thrombus (N = 96) p-Value Odds ratio 95% CI for odds ratio

Bilateral leg complaint 8 45 .053 .43 .18, 1.03

Unilateral left leg complaint 13 45 .923 1.04 .47, 2.30
Unilateral right leg complaint 17 44 .065 2.07 .95, 4.51
Leg edema 29 78 .008 3.09 1.32, 7.22
Leg pain 28 93 .499 1.34 0.58, 3.09
Leg erythema 8 13 .009 4.85 1.47, 15.97
Fever 7 11 .012 5.19 1.42, 18.95
Leg warmth 12 18 <.001 6.92 2.37, 20.24

Statistical method: p-values were calculated based on chi-square or Fisher’s exact tests.

Table 4 focused on answering our main question if we needed to change

Patients in study period classified by cancer site and leg DVT status.
Cancer site Total in study Total with DVT
Cervix 289 16
Ovary/fallopian tube/peritoneal 195 8 4.1%
Uterus 255 10
Vagina 6 0
Vulva 36 1 2.8%
Other 39 1 2.6%
Could we lower this rate with additional heparin? It is possible but
the American College of Obstetrics and Gynecology considers the
practice of combining heparin and SCD as not well supported by
evidence (evidence level C) [10] (Table 4).
Furthermore, our study reaffirms that cancer increases leg DVT
more significantly than surgery. Moreover, the leg DVT rate in our
study using SCD prophylaxis (4.2%) was lower than the first
ENOXACAN study using 1 week of unfractionated heparin (16.5%)
[7]. This difference may be explained by the difference in the study
design. In the first ENOXACAN study, patients were routinely
screened using venography [7]. In our study, we only ordered
Doppler sonogram when we had clinical suspicion for leg DVT. One
could say that screening using venography in non-symptomatic
patients is more accurate. However, the question must be raised as
to the significance of positive venography findings in asympto-
matic patients. Our current practice is to treat asymptomatic
patients found to have thrombus. However, we may overtreat
patients with asymptomatic thrombus if the thrombus forms
transiently with little clinical consequence. What we do not know
is the rate of asymptomatic thrombus found in the general healthy
population. Unfortunately, no large study has identified the
incidence of asymptomatic clot in a general population. The
closest study done in a general population was related to screening
asymptomatic flight passengers where the incidence of asympto-
matic DVT reported a meta-analysis result ranging from 0% to
10.3% [11]. Despite lack of scientific evidence of treating small
asymptomatic thrombus, our current standard of care dictates that
we continue to treat until more rigorous studies can answer this
question definitively.
Our study has some weaknesses. First of all, there were a
number of patients for whom we did not know the final diagnosis
since we sent them back to the referring physicians. However, we
presumed that most if not all of those patients had a benign
diagnosis. These patients are usually referred back to our practice
by their primary physicians if they subsequently are found to have
cancers, since we are the only gynecologic oncology practice in
middle southern United States. Secondly, the study data were
derived from a single gynecologic oncologist, and may not be
applicable to other practices. In the same time, data derived from
this single practice may also decrease variability in data collection
and practice pattern. Thirdly, we did not stratify our study on other
factors that may contribute to development of DVT. Our paper is
our thromboembolic prophylaxis. The other paper has a stratified
Proportion study published showing a higher incidence of thromboembolic
disease in patients who received fresh frozen plasma (19%) than
5.5%
patients who did not (3.3%) [8]. That paper did not find an
3.9% association between age, stage, body mass index, length of the
0% operation, blood loss, presence of ascites, volume of ascites,
residual disease status, preoperative hemoglobin level and
coagulation profile with thromboembolic disease [8].
We think more studies should be done before we change our
current practice of using SCD on postoperative patients to
screening every patient with venography and administrating
low molecular weight heparin for the full 1 month after surgery as
recommended by the ENOXACAN II study [4]. Indeed, the American
College of Obstetrics and Gynecology has recommended SCD as an
effective thromboembolic prophylaxis in high-risk patients based
on Level A evidence-based medicine [10]. Other studies confirmed
SCD as the most cost effective thromboprophylaxis available
[12,13].
Our study also confirms that using clinical symptoms is a
sensitive but not specific method to diagnose leg DVT. However,
these clinical symptoms are helpful in determining which patients
will need further Doppler study to rule out leg DVT. When using
symptoms to direct when to order a Doppler ultrasound study, we
seemed to over order the Doppler ultrasound in order to find the
true leg DVT. Many patients have leg pain that was not caused by
DVT. However, we ordered tests to rule out DVT on patients with
leg pain since we did not want to miss any possible leg DVT, as the
consequence of missing the diagnosis may be dire to patients. We
hope other investigators would confirm our findings.
Conflict of interest
All authors do not have any financial and personal relationships
with other people or organizations that could inappropriately
influence (bias) their work.
Acknowledgment
We thank Londa Schaffer for extensive administrative and
logistic management of the study.
References
[1] David JD. Prevention, diagnosis, and treatment of venous thromboembolic
complications of gynecologic surgery. Am J Obstet Gynecol 2001;184(4):759–
75.
[2] Crandon AJ, Koutts J. Incidence of post-operative deep vein thrombosis in
gynaecological oncology. Aust N Z J Obstet Gynaecol 1983;23(4):216–9.
[3] Clarke-Pearson DL, Delong E, Cynan IS, Soper JT, Creasman WT, Coleman RE. A
controlled trial of two low dose heparin regimens for prevention of post-
operative DVT. Obstet Gynecol 1990;75:684–9.
[4] Bergqvist D, Agnelli G, Cohen AT, Eldor A, Nilsson PEL, Moigne-Amrani A.
Duration of prophylaxis against venous thromboembolism with enoxaparin
after surgery for cancer (ENOXACAN II). N Engl J Med 2002;346(March
(13)):975–80.
176 J.T. Santoso et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 144 (2009) 173–176
[5] Maxwell GL, Synan I, Dodge R, Carroll B, Clarke-Pearson DL. Pneumatic
compression versus low molecular weight heparin in gynecologic oncology
surgery: a randomized trial. Obstet Gynecol 2001;98(6):989–95.
[6] Clark-Pearson DL, DeLong E, Synan IS, Soper JT, Creasman WT, Coleman RE. A
controlled trial of two low-dose heparin regimens for the prevention
of postoperative deep vein thrombosis. Obstet Gynecol 1990;75(4):
684–9.
[7] Enoxacan Study Group. Efficacy and safety of enoxaparin versus unfractio-
nated heparin for prevention of deep thrombosis in elective cancer surgery: a
double-blind randomized multicentre trial with venographic assessment. Br J
Surg 1997;84(8):1099–103.
[8] Abu-Rustum NR, Richard S, Wilton A, et al. Transfusion utilization during
adnexal or peritoneal cancer surgery: effects on symptomatic venous throm-
boembolism and survival. Gynecol Oncol 2005;99(2):320–6.
[9] Tateo S, Mereu L, Salamano S, et al. Ovarian cancer and venous thromboem-
bolic risk. Gynecol Oncol 2005;99(1):119–25.
[10] Prevention of Deep Vein Thrombosis and Pulmonary Embolism. ACOG Practice
Bulletin. Number 84, 2007. http://www.acog.org/publications/educational_-
bulletins/pb084.cfm.
[11] Adi Y, Bayliss S, Rouse A, Taylor RS. The association between air travel and deep
vein thrombosis: systematic review and meta-analysis. BMC Cardiovasc Dis-
ord 2004;4:7.
[12] Maxwell GL, Myers ER, Clarke-Pearson DL. Cost-effectiveness of deep venous
thrombosis prophylaxis in gynecologic oncology surgery. Obstet Gynecol
2000;95:206–14.
[13] Nicolaides A, Goldhaber SZ, Maxwell GL, et al. Cost benefit of intermittent
pneumatic compression for venous thromboembolism prophylaxis in general
surgery. Int Angiol 2008;27(6):500–6.