- Email: [email protected]
Pregnancy and non-pregnancy related ovarian vein thrombosis: Clinical course and outcome
Thrombosis Research 146 (2016) 84–88
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Full Length Article
Pregnancy and non-pregnancy related ovarian vein thrombosis: Clinical course and outcome Amihai Rottenstreich MD a, Nael Da'as MD b, Geffen Kleinstern PhD c, Galia Spectre MD, PhD d, Hagai Amsalem MD e, Yosef Kalish MD a,⁎ a
Department of Hematology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel Internal Medicine D, Hematology Unit, Shaare Zedek Medical Center, Jerusalem, Israel Braun School of Public Health and Community Medicine, Faculty of Medicine of the Hebrew University and Hadassah, Jerusalem, Israel d Department of Hematology, Coagulation Unit, Beilinson Hospital, Rabin Medical Center, Petah Tikva, Israel and Sackler Faculty of Medicine, Tel-Aviv University, Tel Aviv, Israel e Department of Obstetrics and Gynecology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel b c
a r t i c l e
i n f o
Article history: Received 7 July 2016 Received in revised form 17 August 2016 Accepted 1 September 2016 Available online 5 September 2016 Keywords: Anticoagulation Ovarian vein thrombosis Pregnancy Post-partum Thrombophilia
a b s t r a c t Objective: To collect and summarize demographic, clinical, laboratory and radiologic characteristics, as well as management and follow-up data, of patients diagnosed with ovarian vein thrombosis. Methods: A multicenter retrospective review of all patients diagnosed with ovarian vein thrombosis between January 2000 and May 2015 at three university hospitals. Results: Data of 74 women were analyzed. Mean age was 31 ± 9 years. Sixty (81.1%) cases were pregnancyrelated. The presence of at least one underlying risk factor (most commonly active infection or surgery) was more common among pregnancy than non-pregnancy related cases (61.7% vs. 14.3%, P = 0.002). Anticoagulation therapy was administered in 98.6% of patients and adjunctive antibiotic therapy in 39 (52.7%). At a median follow-up of 40 ± 38 months, only one recurrent thrombotic event was observed, and no events of death. Median duration of anticoagulation treatment tended to be longer among patients with non-pregnancy related OVT (6 months [3–14] vs. 3 months [3–6], P = 0.1). Thrombophilic evaluation detected any thrombophilic risk factor in 12 (20%) and 6 (42.9%) women with pregnancy and non-pregnancy related ovarian vein thrombosis, respectively (P = 0.09). Conclusion: Pregnancy-related ovarian vein thrombosis is characterized by a provoked nature and a high rate of resolution after short term treatment. Treatment of three months duration of anticoagulation following this condition appeared in this study to be safe, with no recurrences encountered during a median follow up of 40 months. Thrombophilia seems to have an important role in ovarian vein thrombosis and should be evaluated in non-pregnancy related cases. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Ovarian vein occlusion due to thrombosis was first described in 1909 by Vineberg [1]. Ovarian vein thrombosis (OVT) occurs in 0.05–0.18% of pregnancies; most often encountered in the setting of the postpartum period [2]. Nevertheless, OVT is also associated with other causes such as prior pelvic surgery, malignancy, inflammatory bowel disease, and pelvic inflammatory disease [3–4]. Much of the available data on OVT are based on older publications and scattered case reports [3,5–8]. The clinical course, management strategies, and the role of thrombophilia among patients with OVT are poorly characterized, and conflicting results reported. Moreover, the natural history of OVT, in terms of ⁎ Corresponding author at: Department of Hematology, Hadassah-Hebrew University Medical Center, POB 12000, Jerusalem 91120, Israel. E-mail address: [email protected] (Y. Kalish).
http://dx.doi.org/10.1016/j.thromres.2016.09.001 0049-3848/© 2016 Elsevier Ltd. All rights reserved.
recurrence and expected survival, has not been well-described. Given the paucity of literature regarding clinical presentation, management and outcome of OVT, we performed this multicenter retrospective study of patients diagnosed with OVT in three large university hospitals in Israel. We also aimed to investigate whether the underlying etiology of OVT is related to the clinical course and treatment. 2. Materials and methods 2.1. Data collection Three university hospitals in Israel participated in this study; each maintains an electronic database of all patients diagnosed with OVT. For all the patients included in this study the diagnosis of OVT was recorded in their electronic file. After hospital discharge, a special team of secretaries reviewed patient hospitalization files and added
A. Rottenstreich et al. / Thrombosis Research 146 (2016) 84–88
diagnoses that were not previously recorded. Between August and December 2015, we reviewed the electronic medical records of all cases of OVT managed between January 1, 2000 and May 31, 2015 at three hospitals. Data extracted included patient characteristics, clinical manifestations, laboratory findings, radiologic studies, and management and follow-up data from our thrombosis clinic. OVT was defined as ‘acute’ when all symptoms presented and persisted for b3 weeks; ‘chronic’ when at least one symptom persisted for N 3 weeks; and ‘asymptomatic’ when none of the symptoms were present and OVT was diagnosed incidentally during radiologic evaluation. Thrombophilic evaluation, performed remotely from the thrombotic complication, included Prothrombin 20210, Factor V Leiden, Antithrombin, Protein C and S, anti-phospholipid antibodies and factor VIII. All patients who were diagnosed with antiphospholipid syndrome fulfilled the current diagnostic criteria-Sydney revision of Sapporo criteria [9]. Antiphospholipid antibodies tested included lupus anticoagulant, anticardiolipin IgM and IgG, anti beta2-glycoprotein1 IgM and IgG. The diagnosis of protein S deficiency was accepted only when testing was performed at least twice, from 6 months following pregnancy, following a period of time off anticoagulation. Institutional review board approval waiving informed consent was obtained for this retrospective study (No. HMO 0501-15) from Hadassah Medical Center Helsinki Committee. 2.2. Statistical analysis Patient characteristics are described as means ± SD and medians with interquartile range for continuous variables and as proportions for categorical variables. Significance between pregnancy and nonpregnancy related cases was assessed using the chi-square and Fisher's exact tests for categorical variables, while the Mann-Whitney U test was used for continuous variables without a normal distribution. Results are reported using P-values. A 2-sided P-value b 0.05 indicated statistical significance. SPSS 17.0 for Windows (IBM SPSS, New York, NY) was used for data management and analysis. 3. Results 3.1. Patient characteristics A total of 74 women with OVT were identified, mean age 32 ± 9 years (range 19–68). Sixty (81.1%) cases were pregnancy-related and the remaining 14 (18.9%) were non-pregnancy related. The mean age of the group of patients with pregnancy-related OVT was significantly lower than that of patients with non-pregnancy related OVT (mean 31 ± 6 vs. 40 ± 15 years, P = 0.05). 3.2. Clinical features Most patients (89.2%) had acute presentation, whereas 6.8% and 4.1% had chronic and asymptomatic courses, respectively. The most common symptom was lower abdominal pain (78.4%), followed by constitutional symptoms such as nausea, vomiting, or anorexia (14.9%), and malaise (8.1%). Fever was present in 52.7% of patients. On examination, abdominal tenderness was noted in 57 (77%). Palpable abdominal mass was found in one patient (1.4%). Compared to non-pregnancy related cases, patients with pregnancyrelated OVT tended to present with fever (P = 0.02) and acute symptomatology (acute presentation: 93.3% vs. 71.4%, chronic: 3.3% vs. 21.4%, asymptomatic: 3.3% vs. 7.1%, P = 0.04). Symptoms such as nausea, vomiting, anorexia, and malaise were more common among non-pregnancy related cases (P = 0.004 and P = 0.08, respectively) (Table 1). In 60 (81.1%) patients, OVT occurred in the setting of pregnancy, mostly in the puerperium (n = 54), and the remaining during the first trimester of pregnancy (n = 3), or following first trimester elective
85
abortion (n = 3). The mean number of children among these women was 3.6 (range 0–12). During the study period, the approximate total number of births at the 3 participating hospitals was 500,000, resulting in a rate of 0.01% of puerperal OVT. Among puerperal cases, symptoms appeared after a median of 6 days (range 2–37) following delivery. Most of these cases (85%) occurred following singleton pregnancy, 13.3% following twin pregnancy, and one (1.7%) following triplet pregnancy. The mode of delivery was vaginal and caesarean in 50% of the cases each. The median week of delivery was 38 (range 27–42), and the median birth weight 3050 (range 875–4300) grams. 3.3. Clinical risk factors In 24 patients (32.4%), OVT occurred in the course of an acute infection: lower urinary tract infection (n = 8), Gram positive or negative sepsis (n = 4), endometritis (n = 4), chorioamnionitis (n = 3), pyelonephritis (n = 2), extensive wound infection (n = 1), CMV-positive infectious mononucleosis (n = 1), and Clostridium difficile associated diarrhea (n = 1). Microorganisms that were cultured included Escherichia coli, Proteus mirabilis, Klebsiella pneumonia, Enterococcus, group A streptococci, Staphylococcus aureus, Bacteroides fragilis, and Clostridium difficile. In 33 patients (44.6%), OVT occurred postoperatively after caesarean section (n = 27), total abdominal hysterectomy (n = 2), dilation and curettage (n = 2), small bowel resection (n = 1), and surgical hysteroscopy (n = 1). Evidence of preceding trauma was present in two women (2.7%). Only one patient (1.4%) had active cancer at the time of diagnosis. Nine patients (12.2%) had a personal history of an unprovoked venous thrombotic event (4 pulmonary embolism, 4 deep vein thrombosis, and one mesenteric venous thrombosis); the rate was higher among non-pregnancy related cases (28.6% vs. 8.3%, P = 0.06). In all five pregnancy related cases low molecular weight heparin prophylaxis was administered at the time of diagnosis. The remaining four patients with non-pregnancy related OVT, were not receiving anticoagulation therapy when OVT was diagnosed. At least one underlying risk factor for OVT was found in 37 (61.7%) of pregnancy-related cases compared to 2 (14.3%) non-pregnancy related cases (P = 0.002) (Table 1). 3.4. Laboratory findings Mild anemia was present in 40 (54.1%) patients at presentation. Only 8 (10.8%) presented with thrombocytopenia (platelet count b 150,000/mm3). Elevated white blood cell count was found in 39 (52.7%) patients. A 2–4-fold increase in lactate dehydrogenase was detected in 37.8% of patients. Both of the latter findings were more common in pregnancy- related cases (P = 0.001 and P = 0.01, respectively). Increased inflammatory markers (elevated C-reactive protein or erythrocyte sedimentation rate) were observed in 13 of 25 patients (52%) in whom they were tested. Only one patient (1.4%) had impaired kidney function at the time of diagnosis. 3.5. Diagnosis Diagnostic imaging was performed in all patients. Computed tomography (CT) served as the diagnostic modality in 68 (91.9%). In the remaining six patients, the diagnosis of OVT was confirmed by means of Doppler ultrasonography (n = 4, 5.4%) and magnetic resonance imaging (n = 2, 2.7%). OVT was unilateral in 73 (98.6%) patients, affecting mostly the right side (n = 61, 82.4%). Extension of thrombus beyond the affected ovarian vein occurred in 19 patients (25.7%); and involved the inferior vena cava (n = 7), renal vein (n = 4) and iliac vein (n = 2). Non-fatal pulmonary embolism was diagnosed in association with OVT
86
A. Rottenstreich et al. / Thrombosis Research 146 (2016) 84–88
Table 1 Patients' demographic and clinical characteristics.
Number of patients (%) Age; median [IQR] (mean) Underlying conditiona Active infection Post-surgical Trauma Cancer None Personal history of VTE Presentationb Acute Chronic Asymptomatic Symptoms Abdominal pain Nausea, vomiting Anorexia, malaise Signs Fever Abdominal tenderness Abdominal mass Laboratory findings Anemia Leukocytosis Thrombocytopenia Acute kidney injury Increased LDH Elevated CRP/ESR At least one laboratory abnormality Location of OVT Left Right Bilateral Thrombosis extension Treatment Supportive Anticoagulation Anticoagulation and antibiotics Hospitalization duration (days); median [IQR] (mean) Thrombophilia VTE recurrence Anticoagulation therapy duration (months); median [IQR] (mean)
All patients
Pregnancy-related
Non-pregnancy related
P-value
74 (100%) 31 [27–36] (32)
60 (81.1%) 30 [27–35] (31)
14 (18.9%) 38 [27–51] (40)
– 0.05
24 (32.4%) 33 (44.6%) 2 (2.7%) 1 (1.4%) 35 (47.3%) 9 (12.2%)
22 (36.7%) 31 (51.7%) 2 (3.3%) 0 23 (38.3%) 5 (8.3%)
2 (14.3%) 2 (14.3%) 0 1 (7.1%) 12 (85.7%) 4 (28.6%)
0.13 0.01 1 0.19 0.002 0.06 0.04
66 (89.2%) 5 (6.8%) 3 (4.1%)
56 (93.3%) 2 (3.3%) 2 (3.3%)
10 (71.4%) 3 (21.4%) 1 (7.1%)
58 (78.4%) 11 (14.9%) 6 (8.1%)
46 (76.7%) 5 (8.3%) 3 (5%)
12 (85.7%) 6 (42.9%) 3 (21.4%)
0.72 0.004 0.08
39 (52.7%) 57 (77.0%) 1 (1.4%)
36 (60.0%) 46 (76.7%) 1 (1.7%)
3 (21.4%) 11 (78.6%) 0
0.02 1 1
40 (54.1%) 39 (52.7%) 8 (10.8%) 1 (1.4%) 28 (37.8%) 52%c 65 (87.8%)
33 (55.0%) 36 (60.0%) 5 (8.3%) 0 27 (45%) – 55 (91.7%)
7 (50.0%) 3 (21.4%) 3 (21.4%) 1 (7.1%) 1 (7.1%) – 10 (71.4%)
0.77 0.01 0.17 0.19 0.01 – 0.06
12 (16.2%) 61 (82.4%) 1 (1.4%) 19 (25.7%)
8 (13.3%) 51 (85.0%) 1 (1.7%) 15 (25.0%)
4 (28.6%) 10 (71.4%) 0 4 (28.6%)
0.37
1 (1.4%) 34 (45.9%) 39 (52.7%) 7 [4–14] (10) 18 (24.3%) 1 (1.4%) 3 [3–6] (8)
0 24 (40.0%) 36 (60%) 9 [5–14] (11) 12 (20%) 0 3 [3–6] (5)
1 (7.1%) 10 (71.4%) 3 (21.4%) 4 [2–6] (6) 6 (42.9%) 1 (7.1%) 6 [3–14] (18)
0.75 0.007
0.005 0.09 0.19 0.1
N-number; M-male; F-female; SD-standard deviation; OVT-renal vein thrombosis; LDH-lactate dehydrogenase; CRP-C-reactive protein; ESR-erythrocyte sedimentation rate; VTE-venous thromboembolism. Note- All but three variables are binary: mode of presentation, location of OVT, and treatment. a More than one underlying condition was present in some of the patients. b Acute-duration of symptoms b 3 weeks, chronic-duration of symptoms N 3 weeks. c In 13 of 25 patients who tested CRP and ESR, elevated levels were found.
in 7 (9.5%) patients. Concurrent splenic infarct and renal artery thrombosis were demonstrated in one patient each. OVT was diagnosed incidentally during imaging evaluation in three cases; two pregnant patients following major trauma and one non-pregnant patient during pre-operative evaluation. 3.6. Management Anticoagulation therapy was administered and continued at discharge in 73 (98.6%) patients. Adjunctive antibiotic therapy was used in 39 (52.7%). Patients with non-pregnancy related OVT were less likely to receive antibiotic therapy (P = 0.007). One patient received supportive therapy alone.
among patients with non-pregnancy related OVT (6 [3–14] vs. 3 months [3–6], P = 0.1). Thrombophilic evaluation was performed in all patients; 12 women (20%) with pregnancy related OVT and 6 (42.9%) with nonpregnancy related OVT (P = 0.09) tested positive. The thrombophilia detected were factor V Leiden (n = 6), protein S deficiency (n = 5), antiphospholipid syndrome (n = 4), prothrombin 20210 mutation (n = 4), and protein C deficiency (n = 1). Follow-up imaging evaluation using CT (n = 54) and ultrasonography, combined with Doppler studies (n = 20), demonstrated complete resolution of thrombus in all cases after a median period of 5 (range 3–10) months. The only event of recurrent OVT occurred in a non-pregnant patient with protein S deficiency six years after the first event. No other recurrent events of venous thromboembolism or deaths were encountered during the follow-up period.
3.7. Follow-up data 4. Discussion The mean duration of hospitalization was 10 ± 8 days (range 2–60) and was significantly longer among pregnancy-related cases (11 ± 9 vs. 6 ± 7 days, P = 0.005). Follow-up data were available from our thrombosis clinic for a mean 40 ± 38 (range 2–152) months. Median duration of anticoagulation therapy was 3 months [3–6] and tended to be longer
This 15-year multicenter review identified 74 cases of OVT; most of them were encountered in the postpartum period. Anticoagulation served as the first-line treatment and led to clinical improvement and resolution of the thrombus in all cases. To the best of our knowledge,
A. Rottenstreich et al. / Thrombosis Research 146 (2016) 84–88
this is the largest follow-up study to investigate the natural history and long term outcomes of women with OVT. Clinical suspicion is of utmost importance in OVT diagnosis, due to the wide spectrum of the initial presentation. In the current cohort, fever presented more often in cases associated with pregnancy than in non-pregnancy related cases; constitutional symptoms were more common in the latter. Palpable abdominal mass, which is described in the literature as a classic finding, was noted in only one patient. Laboratory studies were often within normal limits in this study. Abnormal findings include mild anemia, leukocytosis, and thrombocytopenia; modestly increased levels of lactate dehydrogenase and inflammatory markers were sometimes detected. However, none of these tests are pathognomonic. In light of the unspecific clinical presentation and the absence of specific laboratory findings, imaging constitutes the cornerstone of OVT diagnosis. CT, which served as the diagnostic modality in most of our patients, is currently the test of choice for the diagnosis of OVT, with high sensitivity and specificity rates [10,11]. Ultrasonography is less sensitive, as it is incapable of demonstrating the entire course of the ovarian vein, and is significantly limited by patient habitus, bowel gas, and overlying structures [10,12]. Magnetic resonance imaging can also aid in the diagnosis, with high sensitivity rates, and has the advantage of avoiding ionizing radiation. However, magnetic resonance imaging is not readily available and should be reserved for patients with inconclusive CT findings and persistent suspicion of OVT [13,14]. The right side predominance found in our cohort among pregnancyrelated cases is in accordance with previous studies [10]. Dextrotorsion of the enlarged uterus, which compresses the right ovarian vein, has been suggested to account for this finding [10]. Moreover, in the postpartum state, the anterograde flow in the right ovarian vein results in thrombus extension and exposure to a higher bacterial inoculum, whereas the pattern of flow is retrograded in the left ovarian vein [8]. However, in non-pregnant patients, data in the literature are scarce in terms of the predilection to a specific side. In our cohort, right predominance was also noted in this group of patients. The longer course of the right ovarian vein and its multiple incompetent valves probably contribute to its tendency toward thrombosis also in non-pregnant patients [4]. Thrombosis extension beyond the ovarian vein was demonstrated in one quarter of our cohort, with no difference between pregnancy and non-pregnancy related OVT. The most common site of further involvement was the inferior vena cava, occurring in 7 patients (9.5%). Pulmonary embolism was found in 7 patients (9.5%), with no fatal cases. This relatively high rate of thrombosis extension justifies the use of anticoagulation therapy. Our study strongly suggests that pregnancy-related OVT is a highly provoked event. Pregnancy itself is a hypercoagulable state and a risk factor for venous thromboembolism. In addition, active infection, evident by positive bacterial cultures, was present in above one-third of pregnancy-related cases. In fact, rates of active infection may be even higher, considering that almost 60% of pregnancy-related cases presented with fever. Caesarean section and multiple pregnancy are established risk factors for OVT [4,11]. Their rates in our study were significantly higher than their overall rates in Israel (50% vs. 19.5% for caesarean section and 15% vs. 4.5% for multiple pregnancy) [15]. Currently, there is no consensus regarding the optimal duration of anticoagulation therapy [3,4]. Pregnancy-related OVT is characterized by a provoked nature and a high rate of resolution after short term treatment. In this study, treatment of three months duration of anticoagulation appeared to be safe, with no recurrences encountered during a median follow up of 40 months. The high rates of active infection found in our study support the use of broad spectrum antibiotics during the acute phase. In non-pregnancy related cases, a causative or underlying risk factor was mostly absent compared to pregnancy-related cases (P = 0.002). According to our experience, we suggest considering a longer duration
87
of anticoagulation therapy, up to 6 months, especially in patients with thrombosis extension or positive thrombophilic evaluation. Hereditary prothrombotic risk factors seem to play an important role in the development of OVT. Thrombophilic evaluation was performed in all patients in our series; results were positive in 20% of those with pregnancy-related OVT, which is similar to the proportion found in patients with lower extremity deep vein thrombosis in previous studies [11,16]. In non-pregnancy related OVT, thrombophilia appears to have a major role and was present in 43% of patients. This high rate of positive thrombophilic evaluation presumably accounts, at least in part, for the longer duration of anticoagulation therapy in nonpregnancy related cases. Despite the low rates of recurrent thrombosis, we support routine screening for prothrombotic risk factors in patients with OVT, especially in non-pregnancy related cases, to guide management in future high risk conditions (i.e. future pregnancies, long-haul air travel, surgery, hormonal therapy, etc.). An important finding of our study is that recurrent venous thrombosis was extremely rare and no deaths occurred over the study period. This is in accordance with previous studies, in which most events of death and recurrent thrombosis in patients with OVT occurred in those with active malignancy [3,17]. For patients with no evidence of active cancer (only one patient in the current cohort), survival rates are not expected to be affected, and the risk for recurrent thrombosis is very low. The retrospective design raises the possibility of biases inherent to such investigations. Another limitation is the lack of a systematic approach regarding the administration and duration of anticoagulation and antibiotic therapies. Nevertheless, the strengths of the study include its large cohort of patients and its relatively long follow-up period. 5. Conclusions This is the largest study to date documenting the clinical course and outcomes of OVT. OVT is a rare event, which requires high clinical index of suspicion due to its vague presentation. CT should serve as the diagnostic modality in suspected cases. Three months duration of anticoagulation following pregnancy-related OVT appeared in this study to be safe with no recurrences encountered during follow-up period. Adjunctive antibiotic therapy should be considered in the acute initial period in pregnancy related cases, as high rates of active infection were found in this group of patients. Thrombophilic evaluation has an important role in patients with non-pregnancy related OVT. Conflict of interests All authors declare they have no conflict of interest. Statement of human and animal rights The study was conducted in accordance with the standards of the Medical Ethics Committee and with the Helsinki Declaration of 1975, as revised in 2008. Informed consent For this type of study informed consent is not required. Funding No external funding was used in this conduct of this study. Author contributions All authors - study concept, design, review and approval of the final manuscript. GK - data analysis. ND, GS, HA, AR- data acquisition. YK, ARdata analysis and writing of the manuscript.
88
A. Rottenstreich et al. / Thrombosis Research 146 (2016) 84–88
Acknowledgements We thank Ms. Cindy Cohen for her editorial assistance. References [1] H.N. Vineberg, Ligation of pelvic veins for puerperal pyemia, Am. J. Obstet. NY 412 (1909). [2] P. Sharma, S. Abdi, Ovarian vein thrombosis, Clin. Radiol. 67 (2012) 893–898. [3] E.M. Wysokinska, D. Hodge, R.D. McBane, Ovarian vein thrombosis: incidence of recurrent venous thromboembolism and survival, Thromb. Haemost. 96 (2006) 126–131. [4] M.S. Gakhal, H.M. Levy, M. Spina, C. Wrigley, Ovarian vein thrombosis: analysis of patient age, etiology, and side of involvement, Del. Med. J. 85 (2013) 45–50. [5] T. Togan, H. Turan, E. Cifci, C. Ciftci, Ovarian and renal vein thrombosis: a rare cause of fever outer the postpartum period, Case Rep. Obstet. Gynecol. 2015 (2015), 817862 (4 pages). [6] I. Djakovic, M. Mustapic, F. Marleku, O. Grgic, Z. Djakovic, V. Kosec, Ovarian vein thrombosis-a case report, Acta Clin. Belg. 70 (2015) 445–446. [7] F. Mone, G. McKeown, A. B., Ovarian vein thrombosis in pregnancy and the puerperium-a case series, J. Obstet. Gynaecol. 35 (2015) 853–854. [8] D.A. Klima, T.E. Snyder, Postpartum ovarian vein thrombosis, Obstet. Gynecol. 1112 (2008), 431e5.
[9] S. Miyakis, M.D. Lockshin, T. Atsumi, D.W. Branch, R.L. Brey, R. Cervera, et al., International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS), J. Thromb. Haemost. 4 (2006) 295–306. [10] M.A. Kominiarek, J.U. Hibbard, Postpartum ovarian vein thrombosis: an update, Obstet. Gynecol. Surv. 61 (2006) 337e42. [11] O. Salomon, M. Dulitzky, S. Apter, New observations in postpartum ovarian vein thrombosis: experience of single center, Blood Coagul. Fibrinolysis 21 (2010) 16–19. [12] A. Girolami, B. Girolami, M. Treleani, E. Bonamigo, V. Tasinato, Venous thrombosis in rare or unusual sites: a diagnostic challenge, Semin. Thromb. Hemost. 40 (2014) 81–87. [13] D.M. Twickler, A.T. Setiawan, R.S. Evans, W.A. Erdman, R.W. Stettler, C.E. Brown, et al., Imaging of puerperal septic thrombophlebitis: prospective comparison of MR imaging, CT, and sonography, AJR Am. J. Roentgenol. 169 (1997) 1039–1043. [14] K. De Cuyper, M. Eyselbergs, P. Bernard, L. Clabout, F.M. Vanhoenacker, Added value of diffusion weighted MR imaging in the diagnosis of postpartum ovarian vein thrombosis, JBR-BTR 97 (2014) 242–244. [15] Statistical abstract of Israel 2013. State of Israel 2013Available at http://www.cbs. gov.il/reader/cw_usr_view_SHTML?ID=754 (accessed on May 29) 2016. [16] O. Salomon, S. Apter, D. Shaham, N. Hiller, J. Bar-Ziv, Y. Itzchak, et al., Risk factors associated with postpartum ovarian vein thrombosis, Blood Coagul. Fibrinolysis 82 (1999) 1015–1019. [17] N. Labropoulos, R.D. Malgor, M. Comito, A.P. Gasparis, P.J. Pappas, A.K. Tassiopulos, The natural history and treatment outcomes of symptomatic ovarian vein thrombosis, J. Vasc. Surg. Venous Lym. Dis. 3 (2015) 42–47.
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Full Length Article
Pregnancy and non-pregnancy related ovarian vein thrombosis: Clinical course and outcome Amihai Rottenstreich MD a, Nael Da'as MD b, Geffen Kleinstern PhD c, Galia Spectre MD, PhD d, Hagai Amsalem MD e, Yosef Kalish MD a,⁎ a
Department of Hematology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel Internal Medicine D, Hematology Unit, Shaare Zedek Medical Center, Jerusalem, Israel Braun School of Public Health and Community Medicine, Faculty of Medicine of the Hebrew University and Hadassah, Jerusalem, Israel d Department of Hematology, Coagulation Unit, Beilinson Hospital, Rabin Medical Center, Petah Tikva, Israel and Sackler Faculty of Medicine, Tel-Aviv University, Tel Aviv, Israel e Department of Obstetrics and Gynecology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel b c
a r t i c l e
i n f o
Article history: Received 7 July 2016 Received in revised form 17 August 2016 Accepted 1 September 2016 Available online 5 September 2016 Keywords: Anticoagulation Ovarian vein thrombosis Pregnancy Post-partum Thrombophilia
a b s t r a c t Objective: To collect and summarize demographic, clinical, laboratory and radiologic characteristics, as well as management and follow-up data, of patients diagnosed with ovarian vein thrombosis. Methods: A multicenter retrospective review of all patients diagnosed with ovarian vein thrombosis between January 2000 and May 2015 at three university hospitals. Results: Data of 74 women were analyzed. Mean age was 31 ± 9 years. Sixty (81.1%) cases were pregnancyrelated. The presence of at least one underlying risk factor (most commonly active infection or surgery) was more common among pregnancy than non-pregnancy related cases (61.7% vs. 14.3%, P = 0.002). Anticoagulation therapy was administered in 98.6% of patients and adjunctive antibiotic therapy in 39 (52.7%). At a median follow-up of 40 ± 38 months, only one recurrent thrombotic event was observed, and no events of death. Median duration of anticoagulation treatment tended to be longer among patients with non-pregnancy related OVT (6 months [3–14] vs. 3 months [3–6], P = 0.1). Thrombophilic evaluation detected any thrombophilic risk factor in 12 (20%) and 6 (42.9%) women with pregnancy and non-pregnancy related ovarian vein thrombosis, respectively (P = 0.09). Conclusion: Pregnancy-related ovarian vein thrombosis is characterized by a provoked nature and a high rate of resolution after short term treatment. Treatment of three months duration of anticoagulation following this condition appeared in this study to be safe, with no recurrences encountered during a median follow up of 40 months. Thrombophilia seems to have an important role in ovarian vein thrombosis and should be evaluated in non-pregnancy related cases. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Ovarian vein occlusion due to thrombosis was first described in 1909 by Vineberg [1]. Ovarian vein thrombosis (OVT) occurs in 0.05–0.18% of pregnancies; most often encountered in the setting of the postpartum period [2]. Nevertheless, OVT is also associated with other causes such as prior pelvic surgery, malignancy, inflammatory bowel disease, and pelvic inflammatory disease [3–4]. Much of the available data on OVT are based on older publications and scattered case reports [3,5–8]. The clinical course, management strategies, and the role of thrombophilia among patients with OVT are poorly characterized, and conflicting results reported. Moreover, the natural history of OVT, in terms of ⁎ Corresponding author at: Department of Hematology, Hadassah-Hebrew University Medical Center, POB 12000, Jerusalem 91120, Israel. E-mail address: [email protected] (Y. Kalish).
http://dx.doi.org/10.1016/j.thromres.2016.09.001 0049-3848/© 2016 Elsevier Ltd. All rights reserved.
recurrence and expected survival, has not been well-described. Given the paucity of literature regarding clinical presentation, management and outcome of OVT, we performed this multicenter retrospective study of patients diagnosed with OVT in three large university hospitals in Israel. We also aimed to investigate whether the underlying etiology of OVT is related to the clinical course and treatment. 2. Materials and methods 2.1. Data collection Three university hospitals in Israel participated in this study; each maintains an electronic database of all patients diagnosed with OVT. For all the patients included in this study the diagnosis of OVT was recorded in their electronic file. After hospital discharge, a special team of secretaries reviewed patient hospitalization files and added
A. Rottenstreich et al. / Thrombosis Research 146 (2016) 84–88
diagnoses that were not previously recorded. Between August and December 2015, we reviewed the electronic medical records of all cases of OVT managed between January 1, 2000 and May 31, 2015 at three hospitals. Data extracted included patient characteristics, clinical manifestations, laboratory findings, radiologic studies, and management and follow-up data from our thrombosis clinic. OVT was defined as ‘acute’ when all symptoms presented and persisted for b3 weeks; ‘chronic’ when at least one symptom persisted for N 3 weeks; and ‘asymptomatic’ when none of the symptoms were present and OVT was diagnosed incidentally during radiologic evaluation. Thrombophilic evaluation, performed remotely from the thrombotic complication, included Prothrombin 20210, Factor V Leiden, Antithrombin, Protein C and S, anti-phospholipid antibodies and factor VIII. All patients who were diagnosed with antiphospholipid syndrome fulfilled the current diagnostic criteria-Sydney revision of Sapporo criteria [9]. Antiphospholipid antibodies tested included lupus anticoagulant, anticardiolipin IgM and IgG, anti beta2-glycoprotein1 IgM and IgG. The diagnosis of protein S deficiency was accepted only when testing was performed at least twice, from 6 months following pregnancy, following a period of time off anticoagulation. Institutional review board approval waiving informed consent was obtained for this retrospective study (No. HMO 0501-15) from Hadassah Medical Center Helsinki Committee. 2.2. Statistical analysis Patient characteristics are described as means ± SD and medians with interquartile range for continuous variables and as proportions for categorical variables. Significance between pregnancy and nonpregnancy related cases was assessed using the chi-square and Fisher's exact tests for categorical variables, while the Mann-Whitney U test was used for continuous variables without a normal distribution. Results are reported using P-values. A 2-sided P-value b 0.05 indicated statistical significance. SPSS 17.0 for Windows (IBM SPSS, New York, NY) was used for data management and analysis. 3. Results 3.1. Patient characteristics A total of 74 women with OVT were identified, mean age 32 ± 9 years (range 19–68). Sixty (81.1%) cases were pregnancy-related and the remaining 14 (18.9%) were non-pregnancy related. The mean age of the group of patients with pregnancy-related OVT was significantly lower than that of patients with non-pregnancy related OVT (mean 31 ± 6 vs. 40 ± 15 years, P = 0.05). 3.2. Clinical features Most patients (89.2%) had acute presentation, whereas 6.8% and 4.1% had chronic and asymptomatic courses, respectively. The most common symptom was lower abdominal pain (78.4%), followed by constitutional symptoms such as nausea, vomiting, or anorexia (14.9%), and malaise (8.1%). Fever was present in 52.7% of patients. On examination, abdominal tenderness was noted in 57 (77%). Palpable abdominal mass was found in one patient (1.4%). Compared to non-pregnancy related cases, patients with pregnancyrelated OVT tended to present with fever (P = 0.02) and acute symptomatology (acute presentation: 93.3% vs. 71.4%, chronic: 3.3% vs. 21.4%, asymptomatic: 3.3% vs. 7.1%, P = 0.04). Symptoms such as nausea, vomiting, anorexia, and malaise were more common among non-pregnancy related cases (P = 0.004 and P = 0.08, respectively) (Table 1). In 60 (81.1%) patients, OVT occurred in the setting of pregnancy, mostly in the puerperium (n = 54), and the remaining during the first trimester of pregnancy (n = 3), or following first trimester elective
85
abortion (n = 3). The mean number of children among these women was 3.6 (range 0–12). During the study period, the approximate total number of births at the 3 participating hospitals was 500,000, resulting in a rate of 0.01% of puerperal OVT. Among puerperal cases, symptoms appeared after a median of 6 days (range 2–37) following delivery. Most of these cases (85%) occurred following singleton pregnancy, 13.3% following twin pregnancy, and one (1.7%) following triplet pregnancy. The mode of delivery was vaginal and caesarean in 50% of the cases each. The median week of delivery was 38 (range 27–42), and the median birth weight 3050 (range 875–4300) grams. 3.3. Clinical risk factors In 24 patients (32.4%), OVT occurred in the course of an acute infection: lower urinary tract infection (n = 8), Gram positive or negative sepsis (n = 4), endometritis (n = 4), chorioamnionitis (n = 3), pyelonephritis (n = 2), extensive wound infection (n = 1), CMV-positive infectious mononucleosis (n = 1), and Clostridium difficile associated diarrhea (n = 1). Microorganisms that were cultured included Escherichia coli, Proteus mirabilis, Klebsiella pneumonia, Enterococcus, group A streptococci, Staphylococcus aureus, Bacteroides fragilis, and Clostridium difficile. In 33 patients (44.6%), OVT occurred postoperatively after caesarean section (n = 27), total abdominal hysterectomy (n = 2), dilation and curettage (n = 2), small bowel resection (n = 1), and surgical hysteroscopy (n = 1). Evidence of preceding trauma was present in two women (2.7%). Only one patient (1.4%) had active cancer at the time of diagnosis. Nine patients (12.2%) had a personal history of an unprovoked venous thrombotic event (4 pulmonary embolism, 4 deep vein thrombosis, and one mesenteric venous thrombosis); the rate was higher among non-pregnancy related cases (28.6% vs. 8.3%, P = 0.06). In all five pregnancy related cases low molecular weight heparin prophylaxis was administered at the time of diagnosis. The remaining four patients with non-pregnancy related OVT, were not receiving anticoagulation therapy when OVT was diagnosed. At least one underlying risk factor for OVT was found in 37 (61.7%) of pregnancy-related cases compared to 2 (14.3%) non-pregnancy related cases (P = 0.002) (Table 1). 3.4. Laboratory findings Mild anemia was present in 40 (54.1%) patients at presentation. Only 8 (10.8%) presented with thrombocytopenia (platelet count b 150,000/mm3). Elevated white blood cell count was found in 39 (52.7%) patients. A 2–4-fold increase in lactate dehydrogenase was detected in 37.8% of patients. Both of the latter findings were more common in pregnancy- related cases (P = 0.001 and P = 0.01, respectively). Increased inflammatory markers (elevated C-reactive protein or erythrocyte sedimentation rate) were observed in 13 of 25 patients (52%) in whom they were tested. Only one patient (1.4%) had impaired kidney function at the time of diagnosis. 3.5. Diagnosis Diagnostic imaging was performed in all patients. Computed tomography (CT) served as the diagnostic modality in 68 (91.9%). In the remaining six patients, the diagnosis of OVT was confirmed by means of Doppler ultrasonography (n = 4, 5.4%) and magnetic resonance imaging (n = 2, 2.7%). OVT was unilateral in 73 (98.6%) patients, affecting mostly the right side (n = 61, 82.4%). Extension of thrombus beyond the affected ovarian vein occurred in 19 patients (25.7%); and involved the inferior vena cava (n = 7), renal vein (n = 4) and iliac vein (n = 2). Non-fatal pulmonary embolism was diagnosed in association with OVT
86
A. Rottenstreich et al. / Thrombosis Research 146 (2016) 84–88
Table 1 Patients' demographic and clinical characteristics.
Number of patients (%) Age; median [IQR] (mean) Underlying conditiona Active infection Post-surgical Trauma Cancer None Personal history of VTE Presentationb Acute Chronic Asymptomatic Symptoms Abdominal pain Nausea, vomiting Anorexia, malaise Signs Fever Abdominal tenderness Abdominal mass Laboratory findings Anemia Leukocytosis Thrombocytopenia Acute kidney injury Increased LDH Elevated CRP/ESR At least one laboratory abnormality Location of OVT Left Right Bilateral Thrombosis extension Treatment Supportive Anticoagulation Anticoagulation and antibiotics Hospitalization duration (days); median [IQR] (mean) Thrombophilia VTE recurrence Anticoagulation therapy duration (months); median [IQR] (mean)
All patients
Pregnancy-related
Non-pregnancy related
P-value
74 (100%) 31 [27–36] (32)
60 (81.1%) 30 [27–35] (31)
14 (18.9%) 38 [27–51] (40)
– 0.05
24 (32.4%) 33 (44.6%) 2 (2.7%) 1 (1.4%) 35 (47.3%) 9 (12.2%)
22 (36.7%) 31 (51.7%) 2 (3.3%) 0 23 (38.3%) 5 (8.3%)
2 (14.3%) 2 (14.3%) 0 1 (7.1%) 12 (85.7%) 4 (28.6%)
0.13 0.01 1 0.19 0.002 0.06 0.04
66 (89.2%) 5 (6.8%) 3 (4.1%)
56 (93.3%) 2 (3.3%) 2 (3.3%)
10 (71.4%) 3 (21.4%) 1 (7.1%)
58 (78.4%) 11 (14.9%) 6 (8.1%)
46 (76.7%) 5 (8.3%) 3 (5%)
12 (85.7%) 6 (42.9%) 3 (21.4%)
0.72 0.004 0.08
39 (52.7%) 57 (77.0%) 1 (1.4%)
36 (60.0%) 46 (76.7%) 1 (1.7%)
3 (21.4%) 11 (78.6%) 0
0.02 1 1
40 (54.1%) 39 (52.7%) 8 (10.8%) 1 (1.4%) 28 (37.8%) 52%c 65 (87.8%)
33 (55.0%) 36 (60.0%) 5 (8.3%) 0 27 (45%) – 55 (91.7%)
7 (50.0%) 3 (21.4%) 3 (21.4%) 1 (7.1%) 1 (7.1%) – 10 (71.4%)
0.77 0.01 0.17 0.19 0.01 – 0.06
12 (16.2%) 61 (82.4%) 1 (1.4%) 19 (25.7%)
8 (13.3%) 51 (85.0%) 1 (1.7%) 15 (25.0%)
4 (28.6%) 10 (71.4%) 0 4 (28.6%)
0.37
1 (1.4%) 34 (45.9%) 39 (52.7%) 7 [4–14] (10) 18 (24.3%) 1 (1.4%) 3 [3–6] (8)
0 24 (40.0%) 36 (60%) 9 [5–14] (11) 12 (20%) 0 3 [3–6] (5)
1 (7.1%) 10 (71.4%) 3 (21.4%) 4 [2–6] (6) 6 (42.9%) 1 (7.1%) 6 [3–14] (18)
0.75 0.007
0.005 0.09 0.19 0.1
N-number; M-male; F-female; SD-standard deviation; OVT-renal vein thrombosis; LDH-lactate dehydrogenase; CRP-C-reactive protein; ESR-erythrocyte sedimentation rate; VTE-venous thromboembolism. Note- All but three variables are binary: mode of presentation, location of OVT, and treatment. a More than one underlying condition was present in some of the patients. b Acute-duration of symptoms b 3 weeks, chronic-duration of symptoms N 3 weeks. c In 13 of 25 patients who tested CRP and ESR, elevated levels were found.
in 7 (9.5%) patients. Concurrent splenic infarct and renal artery thrombosis were demonstrated in one patient each. OVT was diagnosed incidentally during imaging evaluation in three cases; two pregnant patients following major trauma and one non-pregnant patient during pre-operative evaluation. 3.6. Management Anticoagulation therapy was administered and continued at discharge in 73 (98.6%) patients. Adjunctive antibiotic therapy was used in 39 (52.7%). Patients with non-pregnancy related OVT were less likely to receive antibiotic therapy (P = 0.007). One patient received supportive therapy alone.
among patients with non-pregnancy related OVT (6 [3–14] vs. 3 months [3–6], P = 0.1). Thrombophilic evaluation was performed in all patients; 12 women (20%) with pregnancy related OVT and 6 (42.9%) with nonpregnancy related OVT (P = 0.09) tested positive. The thrombophilia detected were factor V Leiden (n = 6), protein S deficiency (n = 5), antiphospholipid syndrome (n = 4), prothrombin 20210 mutation (n = 4), and protein C deficiency (n = 1). Follow-up imaging evaluation using CT (n = 54) and ultrasonography, combined with Doppler studies (n = 20), demonstrated complete resolution of thrombus in all cases after a median period of 5 (range 3–10) months. The only event of recurrent OVT occurred in a non-pregnant patient with protein S deficiency six years after the first event. No other recurrent events of venous thromboembolism or deaths were encountered during the follow-up period.
3.7. Follow-up data 4. Discussion The mean duration of hospitalization was 10 ± 8 days (range 2–60) and was significantly longer among pregnancy-related cases (11 ± 9 vs. 6 ± 7 days, P = 0.005). Follow-up data were available from our thrombosis clinic for a mean 40 ± 38 (range 2–152) months. Median duration of anticoagulation therapy was 3 months [3–6] and tended to be longer
This 15-year multicenter review identified 74 cases of OVT; most of them were encountered in the postpartum period. Anticoagulation served as the first-line treatment and led to clinical improvement and resolution of the thrombus in all cases. To the best of our knowledge,
A. Rottenstreich et al. / Thrombosis Research 146 (2016) 84–88
this is the largest follow-up study to investigate the natural history and long term outcomes of women with OVT. Clinical suspicion is of utmost importance in OVT diagnosis, due to the wide spectrum of the initial presentation. In the current cohort, fever presented more often in cases associated with pregnancy than in non-pregnancy related cases; constitutional symptoms were more common in the latter. Palpable abdominal mass, which is described in the literature as a classic finding, was noted in only one patient. Laboratory studies were often within normal limits in this study. Abnormal findings include mild anemia, leukocytosis, and thrombocytopenia; modestly increased levels of lactate dehydrogenase and inflammatory markers were sometimes detected. However, none of these tests are pathognomonic. In light of the unspecific clinical presentation and the absence of specific laboratory findings, imaging constitutes the cornerstone of OVT diagnosis. CT, which served as the diagnostic modality in most of our patients, is currently the test of choice for the diagnosis of OVT, with high sensitivity and specificity rates [10,11]. Ultrasonography is less sensitive, as it is incapable of demonstrating the entire course of the ovarian vein, and is significantly limited by patient habitus, bowel gas, and overlying structures [10,12]. Magnetic resonance imaging can also aid in the diagnosis, with high sensitivity rates, and has the advantage of avoiding ionizing radiation. However, magnetic resonance imaging is not readily available and should be reserved for patients with inconclusive CT findings and persistent suspicion of OVT [13,14]. The right side predominance found in our cohort among pregnancyrelated cases is in accordance with previous studies [10]. Dextrotorsion of the enlarged uterus, which compresses the right ovarian vein, has been suggested to account for this finding [10]. Moreover, in the postpartum state, the anterograde flow in the right ovarian vein results in thrombus extension and exposure to a higher bacterial inoculum, whereas the pattern of flow is retrograded in the left ovarian vein [8]. However, in non-pregnant patients, data in the literature are scarce in terms of the predilection to a specific side. In our cohort, right predominance was also noted in this group of patients. The longer course of the right ovarian vein and its multiple incompetent valves probably contribute to its tendency toward thrombosis also in non-pregnant patients [4]. Thrombosis extension beyond the ovarian vein was demonstrated in one quarter of our cohort, with no difference between pregnancy and non-pregnancy related OVT. The most common site of further involvement was the inferior vena cava, occurring in 7 patients (9.5%). Pulmonary embolism was found in 7 patients (9.5%), with no fatal cases. This relatively high rate of thrombosis extension justifies the use of anticoagulation therapy. Our study strongly suggests that pregnancy-related OVT is a highly provoked event. Pregnancy itself is a hypercoagulable state and a risk factor for venous thromboembolism. In addition, active infection, evident by positive bacterial cultures, was present in above one-third of pregnancy-related cases. In fact, rates of active infection may be even higher, considering that almost 60% of pregnancy-related cases presented with fever. Caesarean section and multiple pregnancy are established risk factors for OVT [4,11]. Their rates in our study were significantly higher than their overall rates in Israel (50% vs. 19.5% for caesarean section and 15% vs. 4.5% for multiple pregnancy) [15]. Currently, there is no consensus regarding the optimal duration of anticoagulation therapy [3,4]. Pregnancy-related OVT is characterized by a provoked nature and a high rate of resolution after short term treatment. In this study, treatment of three months duration of anticoagulation appeared to be safe, with no recurrences encountered during a median follow up of 40 months. The high rates of active infection found in our study support the use of broad spectrum antibiotics during the acute phase. In non-pregnancy related cases, a causative or underlying risk factor was mostly absent compared to pregnancy-related cases (P = 0.002). According to our experience, we suggest considering a longer duration
87
of anticoagulation therapy, up to 6 months, especially in patients with thrombosis extension or positive thrombophilic evaluation. Hereditary prothrombotic risk factors seem to play an important role in the development of OVT. Thrombophilic evaluation was performed in all patients in our series; results were positive in 20% of those with pregnancy-related OVT, which is similar to the proportion found in patients with lower extremity deep vein thrombosis in previous studies [11,16]. In non-pregnancy related OVT, thrombophilia appears to have a major role and was present in 43% of patients. This high rate of positive thrombophilic evaluation presumably accounts, at least in part, for the longer duration of anticoagulation therapy in nonpregnancy related cases. Despite the low rates of recurrent thrombosis, we support routine screening for prothrombotic risk factors in patients with OVT, especially in non-pregnancy related cases, to guide management in future high risk conditions (i.e. future pregnancies, long-haul air travel, surgery, hormonal therapy, etc.). An important finding of our study is that recurrent venous thrombosis was extremely rare and no deaths occurred over the study period. This is in accordance with previous studies, in which most events of death and recurrent thrombosis in patients with OVT occurred in those with active malignancy [3,17]. For patients with no evidence of active cancer (only one patient in the current cohort), survival rates are not expected to be affected, and the risk for recurrent thrombosis is very low. The retrospective design raises the possibility of biases inherent to such investigations. Another limitation is the lack of a systematic approach regarding the administration and duration of anticoagulation and antibiotic therapies. Nevertheless, the strengths of the study include its large cohort of patients and its relatively long follow-up period. 5. Conclusions This is the largest study to date documenting the clinical course and outcomes of OVT. OVT is a rare event, which requires high clinical index of suspicion due to its vague presentation. CT should serve as the diagnostic modality in suspected cases. Three months duration of anticoagulation following pregnancy-related OVT appeared in this study to be safe with no recurrences encountered during follow-up period. Adjunctive antibiotic therapy should be considered in the acute initial period in pregnancy related cases, as high rates of active infection were found in this group of patients. Thrombophilic evaluation has an important role in patients with non-pregnancy related OVT. Conflict of interests All authors declare they have no conflict of interest. Statement of human and animal rights The study was conducted in accordance with the standards of the Medical Ethics Committee and with the Helsinki Declaration of 1975, as revised in 2008. Informed consent For this type of study informed consent is not required. Funding No external funding was used in this conduct of this study. Author contributions All authors - study concept, design, review and approval of the final manuscript. GK - data analysis. ND, GS, HA, AR- data acquisition. YK, ARdata analysis and writing of the manuscript.
88
A. Rottenstreich et al. / Thrombosis Research 146 (2016) 84–88
Acknowledgements We thank Ms. Cindy Cohen for her editorial assistance. References [1] H.N. Vineberg, Ligation of pelvic veins for puerperal pyemia, Am. J. Obstet. NY 412 (1909). [2] P. Sharma, S. Abdi, Ovarian vein thrombosis, Clin. Radiol. 67 (2012) 893–898. [3] E.M. Wysokinska, D. Hodge, R.D. McBane, Ovarian vein thrombosis: incidence of recurrent venous thromboembolism and survival, Thromb. Haemost. 96 (2006) 126–131. [4] M.S. Gakhal, H.M. Levy, M. Spina, C. Wrigley, Ovarian vein thrombosis: analysis of patient age, etiology, and side of involvement, Del. Med. J. 85 (2013) 45–50. [5] T. Togan, H. Turan, E. Cifci, C. Ciftci, Ovarian and renal vein thrombosis: a rare cause of fever outer the postpartum period, Case Rep. Obstet. Gynecol. 2015 (2015), 817862 (4 pages). [6] I. Djakovic, M. Mustapic, F. Marleku, O. Grgic, Z. Djakovic, V. Kosec, Ovarian vein thrombosis-a case report, Acta Clin. Belg. 70 (2015) 445–446. [7] F. Mone, G. McKeown, A. B., Ovarian vein thrombosis in pregnancy and the puerperium-a case series, J. Obstet. Gynaecol. 35 (2015) 853–854. [8] D.A. Klima, T.E. Snyder, Postpartum ovarian vein thrombosis, Obstet. Gynecol. 1112 (2008), 431e5.
[9] S. Miyakis, M.D. Lockshin, T. Atsumi, D.W. Branch, R.L. Brey, R. Cervera, et al., International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS), J. Thromb. Haemost. 4 (2006) 295–306. [10] M.A. Kominiarek, J.U. Hibbard, Postpartum ovarian vein thrombosis: an update, Obstet. Gynecol. Surv. 61 (2006) 337e42. [11] O. Salomon, M. Dulitzky, S. Apter, New observations in postpartum ovarian vein thrombosis: experience of single center, Blood Coagul. Fibrinolysis 21 (2010) 16–19. [12] A. Girolami, B. Girolami, M. Treleani, E. Bonamigo, V. Tasinato, Venous thrombosis in rare or unusual sites: a diagnostic challenge, Semin. Thromb. Hemost. 40 (2014) 81–87. [13] D.M. Twickler, A.T. Setiawan, R.S. Evans, W.A. Erdman, R.W. Stettler, C.E. Brown, et al., Imaging of puerperal septic thrombophlebitis: prospective comparison of MR imaging, CT, and sonography, AJR Am. J. Roentgenol. 169 (1997) 1039–1043. [14] K. De Cuyper, M. Eyselbergs, P. Bernard, L. Clabout, F.M. Vanhoenacker, Added value of diffusion weighted MR imaging in the diagnosis of postpartum ovarian vein thrombosis, JBR-BTR 97 (2014) 242–244. [15] Statistical abstract of Israel 2013. State of Israel 2013Available at http://www.cbs. gov.il/reader/cw_usr_view_SHTML?ID=754 (accessed on May 29) 2016. [16] O. Salomon, S. Apter, D. Shaham, N. Hiller, J. Bar-Ziv, Y. Itzchak, et al., Risk factors associated with postpartum ovarian vein thrombosis, Blood Coagul. Fibrinolysis 82 (1999) 1015–1019. [17] N. Labropoulos, R.D. Malgor, M. Comito, A.P. Gasparis, P.J. Pappas, A.K. Tassiopulos, The natural history and treatment outcomes of symptomatic ovarian vein thrombosis, J. Vasc. Surg. Venous Lym. Dis. 3 (2015) 42–47.