A "less is more" approach to breast cancer surgery; Shedding light on vitamin D & CLL prognosis

A "less is more" approach to breast cancer surgery

Over the past several decades, there has been a welcome trend towards less extreme surgery for women with breast cancer. The trend has continued with recently published data about axillary node dissection.

Historically, the rationale for complete axillary lymph node (ALN) dissection was detection and removal of metastases in the ALNs, a result that has important prognostic and treatment implications. But full axillary lymph node dissection (ALND) following sentinel lymph node (SLN) sampling appears to be headed toward becoming a practice of the past.-- and avoidance of ALND should result in fewer surgical complications (e.g., lymphedema, seroma, infection, and pain), increased preservation of shoulder mobility (http://journals.lww.com/oncology-times/Fulltext/2011/03100/My_Take_on_ACOSOG_Z0011_Axillary_Dissection_vs_No.5.aspx) and improved cosmetic appearance, and better self-image of patients.

In a recently published report of the ACOSOG Z011 Phase 3 clinical trial of patients with limited SLN metastatic breast cancer conducted from May 1999 to December 2004 at 115 sites in the U.S., Guiliano and colleagues demonstrated that ALND after SLN dissection (SLND) is unnecessary for good local control and excellent survival outcomes

(http://jama.ama-assn.org/content/305/6/569.full).

Almost 900 patients with early-stage invasive breast cancer who were enrolled in the trial from May 1999 to December 2004 were randomized to undergo lumpectomy, ALND (excision of ≥10 nodes), and tangential whole-breast radiation therapy (TWBRT); or lumpectomy, SLND, and TWBRT. Use of adjuvant systemic therapy (endocrine therapy and/or chemotherapy) was at the discretion of the treating medical oncologist. Approximately 96% of the ALND-treated patients and 97% of the SLND-treated patients received adjuvant systemic therapy. Median follow-up was 6.3 years.

Few recurrences occurred in the axillary nodes of patients in either group. No differences were seen in the low incidence of local and regional recurrences in the patient groups. Furthermore, patients in both groups experienced similar rates of overall survival (OS) and disease-free survival (DFS). Five-year OS was 91.8% and 92.5% in the ALND and SLND groups, respectively. Five-year DFS was 82.2% and 83.9% in patients who received ALND versus SLND.

Thus, for patients who have good prognostic indicators for breast cancers that are treated with a combination of lumpectomy, radiation therapy, and systemic therapy, good clinical outcomes may be obtained without surgical excision of involved axillary nodes. This important finding confirms the efficacy of breast-conserving surgery with adjuvant therapy, representing excellent news for women with breast cancer. Each year, thousands of women newly diagnosed with early-stage breast cancer may benefit from the high survival rates and improved quality of life provided by multi-modality therapy involving conservative surgery.

Copyright 2011 A. Richard Adrouny, M.D., F.A.C.P.

Shedding light on vitamin D & CLL prognosis

Correcting inadequate levels of vitamin D makes sense in treatment of osteoporosis, but what about cancer?

A recent article on vitamin D insufficiency and inferior prognosis of chronic lymphocytic leukemia (CLL) by Shanafelt and colleagues (http://bloodjournal.hematologylibrary.org/content/117/5/1492)is the first evidence that vitamin D deficiency is an independent risk factor in CLL and may represent the first publication that demonstrates such an association in any hematological malignancy.

Approximately 30.5% (119/390) of newly diagnosed CLL patients and almost 40% (61/153) of previously untreated CLL patients in the study were deficient in vitamin D. Following a median of 3 and 9.9 years of follow-up, respectively, the vitamin D-deficient patients in both the newly diagnosed CLL group and the CLL validation group experienced shorter time-to-treatment (TTT) and overall survival (OS).

In this provocative study, vitamin D insufficiency was associated with poorer prognosis for TTT and OS, even in early-stage CLL patients managed by a watch-and-wait approach.Based on analysis of data from all patients, adjusting for age, gender, Rai stage, status of various markers, and presence of cytogenetic abnormalities, vitamin D insufficiency was an independent predictor of TTT. Thus, vitamin D levels in serum may represent an important prognostic marker for CLL.

Because CLL currently is incurable, it would be tempting to hope, as Pepper and Fegan speculate, that normalization of vitamin D might be beneficial for CLL patients(http://bloodjournal.hematologylibrary.org/content/117/5/1439.full),and/or that vitamin D supplementation might have a chemopreventative effect on CLL incidence. Unfortunately, being cautiously optimistic may be too sunny a view at present since Shanafelt and colleagues did not study the therapeutic effects of vitamin D supplementation in CLL patients.

What about the role of vitamin D in preventing or improving outcomes in other types of cancer? Multiple, relevant mechanisms of action of vitamin D in preclinical studies on cancer abound (http://bloodjournal.hematologylibrary.org/content/117/5/1439.full).

According to a recent meta-analysis by an Institute of Medicine (IOM) panel, however, the inconsistent research on vitamin D insufficiency and incidence and mortality of many types of solid tumors and non-Hodgkin's lymphoma - as well as that of cardiovascular disease, diabetes, and autoimmune diseases - does not prove a cause-effect relationship (www.nejm.org/doi/full/10.1056/NeJMp1102022). Most past research has been based on lab, ecologic, and observational studies of vitamin D levels in serum, whereas no completed large-scale randomized clinical trials of vitamin D intake have included cancer as the pre-specified, primary endpoint. Another reason for the difficulty in interpreting prior clinical studies on cancer epidemiology is that vitamin D deficiency has been associated with known cancer risk factors, such as obesity, insufficient physical activity, dark skin pigmentation, and dietary habits.

Five to six years from now, data from ongoing clinical trials of moderate to high doses of vitamin D for chemoprevention of cancer will be available(Chustecka, 3/31/11, www.medscape.com; subscription required). Also, many clinical studies of vitamin D and its derivatives as therapeutics for a variety of cancers are in progress. In a best-case scenario, perhaps eventually we will be able to add prevention or treatment of some type of cancer as another rationale for vitamin D supplementation in certain patients. On the other hand, without compelling evidence of beneficial effect in other diseases, osteoporosis and related bone conditions remain the bona fide reasons for normalizing vitamin D.

Copyright 2011 A. Richard Adrouny, M.D., F.A.C.P.

Non-hormonal, non-surgical therapy for menorrhagia; new drug for renal carcinoma; new thrombophilia

FDA approves tranexamic acid for menorrhagia

The U.S. FDA approved tranexamic acid tablets (Lysteda, Xanodyne Pharmaceuticals), the first non hormonal product cleared to treat menorrhagia.

Tranexamic acid was first approved by the FDA in 1986 as an injection, under the brand name Cyklokapron, and is used to reduce or prevent bleeding during and following tooth extraction in patients with hemophilia.
 It is a synthetic lysine derivative that exerts its antifibrinolytic effect by reversibly blocking lysine binding sites on plasminogen and thus preventing fibrin degradation.

The most common adverse reactions reported in clinical trials included headache, sinus and nasal symptoms, back pain, abdominal pain, muscle and joint pain, muscle cramps, anemia, and fatigue. There was a statistically significant reduction in menstrual blood loss in women who received Lysteda, compared with those taking placebo. Use of Lysteda while taking hormonal contraceptives may increase the risk of blood clots, stroke, or heart attack.

FDA approval of pazopanib for advanced renal cell cancer

Another drug, pazopanib (Votrient, GSK) has been added to the growing menu of options for treatment of advanced renal cell cancer, a disease for which only a few years ago we had few treatment choices. The approval was based on a phase III trial of 435 patients with advanced renal cell cancer who were randomized to either the drug or placebo. The response rate was 30% in the treated group, compared to 3% in the placebo group. The treated group did not show progression of their disease for a median duration of 11 months, compared to 2.8 months for the placebo group.

Pazopanib is an oral drug, and works by inhibiting angiogenesis, which is believed to play an essential role in the growth and spread of malignant tumors.

This is the sixth renal carcinoma drug approved since 2005. (The others are sorafenib (Nexavar), sunitinib (Sutent), temsirolimus (Torisel), everolimus (Afinitor) and bevacizumab (Avastin). All of these drugs are considered “targeted therapies” in contrast to conventional chemotherapy drugs. They each have a specific molecular target. None of these drugs have yet been studied in direct comparison with each other. Other approved drugs for advanced renal cell carcinoma include interleukin 2 and interferon. Trials of combinations of these agents, as well comparison studies of drugs head to head, are likely to be reported in the future.

New form of genetic thrombophilia reported

You can now add Factor IX Padua to your list of thrombophilic disorders. This gain of function genetic variant of Factor IX was identified in a patient with venous thromboembolism by physicians in--you guessed it--Padua, Italy. The subject patient was found to have nearly 800% of normal activity of factor IX. The genetic defect that he carried is a point mutation in the factor IX gene (G31134T transversion) that caused a substitution of leucine for arginine at position 338. Other family members were found to have varying degrees of abnormality, not nearly as great as the proband.

Since Factor IX activity elevation is in the differential of thrombophilic disorders, it seems possible that this specific abnormality could be added in the future to the laboratory evaluation of thrombophilia. Studies will have to be carried out to estimate the frequency of this abnormality in the general population. http://content.nejm.org/cgi/content/full/361/17/1671

Tamoxifen pharmacogenetics; GI cancers aren't sexy; vascular complications of splenectomy

Tamoxifen pharmacogenetics affects clinical outcomes

There is a growing list of drugs whose metabolism (and clinical effects) have been shown to be influenced by genetic mutations. (See my posts of March 1 and March 15, 2009: http://adrounysheme-o-gram.blogspot.com/2009_03_01_archive.html).

Besides important drugs like warfarin and clopidgorel (Plavix), tamoxifen effect has also been shown to be influenced by mutations of a metabolizing enzyme, in this case, cytochrome P450 2D6 (CYP2D6) enzyme.

The Oct 7 2009 issue of JAMA (http://jama.ama-assn.org/cgi/content/full/302/13/1429) features a German-American collaboration on clinical outcomes with tamoxifen stratified according to metabolizer type. Tamoxifen has been a foundation of breast cancer treatment for the last 30 years. Its growth inhibitory effect on breast cancer is mediated by metabolites 4-hydroxytamoxifen and endoxifen. The formation of these active metabolites is catalyzed by CP2D6. One hundred variants of CYP2D6 have been identified with four distinct phenotypes of metabolism: extensive (normal activity); intermediate (reduced activity), poor (no activity) and ultrarapid (high activity). A gene-dose effect has been demonstrated.

The study included almost fourteen hundred women from Germany and the U.S. with stages I, II and III who were given tamoxifen according to standard practices and who underwent pharmacogenetic analysis. Endpoints included standard oncologic parameters, including disease free survival and overall survival. Median follow-up was over six years.

The best results were seen in the extensive (normal activity) metabolizers. The overall recurrence rate was nearly double in the poor metabolizer group (24% vs 12.5%) and the death rate was significantly higher in the poor metabolizer group (22.8% vs. 16.7%)

This study is not the first to suggest an association between CYP2D6 mutations and clinical outcomes. However, it is the first with enough statistical power to actually prove the relationship.

The use of CYP2D6 testing is not yet a clinical standard. It may well become one. I have observed that at least one insurance company has been routinely demanding testing of patients who are prescribed tamoxifen.

An Interesting Comment

“For lack of a better word, GI cancers aren’t particularly sexy, certainly not from a political standpoint, nor is it easy to attract the blandishments of corporate largesse when the victims of these diseases comprise such a nebulous consumer base. Thus…it really falls on our shoulders to inspire and lead our patients and advocacy groups in an effort to increase funding, improve clinical trial participation, or incite a bit of a ‘riot’, if that’s what it takes to shatter insouciance and ensure that GI cancers get the attention they deserve…I strongly believe we can change the outlooks of patients with these terrible diseases…Our patients deserve better than run-of-the-mill interventions and should be demanding optimum treatment.”—John L. Marshall, M.D., Chief, Division of Hematology/Oncology, Lombardi Comprehensive Cancer Center, Georgetown University. Quoted in Gastrointestinal Cancer Research Volume 3 Issue 4.

Splenectomy induces a hypercoagulable state

Over 20,000 splenectomies are performed annually in the U.S. Besides a higher incidence of overwhelming sepsis, there are concerns about other potential long term complications of splenectomy. A recent article in Blood http://bloodjournal.hematologylibrary.org/cgi/content/full/114/14/2861 reviews the possibility that splenectomy induces a higher incidence of vascular complications.

Spleen contains white pulp and red pulp. White pulp is largely lymphoid tissue that processes antigens and produces antibodies. Red pulp consists of the sinusoidal cords of Billroth, which filter the blood by removing aged and damaged cells and by “polishing” red cells that have developed surface imperfections. The red pulp also contains a reservoir of platelets and granulocytes.

A number of potential vascular complications have been identified in post-splenectomy patients. These include arterial events such as stoke, M.I., carotid disease and peripheral arterial disease.

Venous events, both local and systemic, have been identified. The incidence of portal vein thrombosis after splenectomy ranges from 5 to 37%. Fatal pulmonary embolism was 5 times higher in persons with previous splenectomy than in matched controls in one autopsy study.

Pulmonary arterial hypertension is another potential vascular complication.

The authors speculate that the higher incidence of vascular events has multifactorial causes, resulting from hypercoagulability, platelet activation, disturbance and activation of the endothelium and altered lipid profiles. Absence of the filtration function of the spleen may permit particulates and cell debris to activate vascular endothelium. Splenectomy increases platelet counts, cholesterol, C-reactive protein, white count and hemoglobin concentration, all of which are associated with increased risks of arterial and venous thrombosis.

The authors conclude that further study of this is warranted, particularly with respect to issues such as whether short or long term thromboprophylaxis should be offered patients who have had splenectomy.

An alternative to warfarin for A. fib? ; The Great Influenza

Dabigatran for atrial fibrillation

The NEJM recently published a lead article (http://content.nejm.org/cgi/content/full/361/12/1139) which compared dabigatran to warfarin in patients with atrial fibrillation. Dabigatran belongs to a new class of oral anticoagulants known as DTI’s (direct thrombin inhibitors) which are in development. (See my Heme-O-Gram of Dec. 14 ’08 http://adrounysheme-o-gram.blogspot.com/2009/03/new-oral-anticoagulant-drugs-report.html )

While the effectiveness of warfarin for stroke prevention in atrial fibrillation is in little doubt, warfarin is probably underutilized because of a number of reasons primarily relating to inconvenience and risk of bleeding. Drugs like dabigatran offer several advantages over warfarin: fixed dose, shorter half-life, no need for monitoring, wide therapeutic index, oral administration, general convenience. The question is whether it meets non-inferiority with a tried and true agent like warfarin. Previously, dabigatran was shown to be as effective and safe as enoxaparin (Lovenox) in reducing VTE after orthopedic surgery.

The RE-LY (Randomized Evaluation of Long-Term Anticoagulation) trial, which accrued and randomized over 18,000 patients, the standard of non-inferiority for atrial fibrillation appears to have been met with respect to the primary outcome, either stroke or systemic embolism. The study was funded by Boehringer Ingelheim, the manufacturer of dabigatran. Two doses of dabigatran were studied: 110 mg and 150 mg each given b.i.d. The 18,000 patients were evenly matched across the three study groups.

The lowest rate of stroke/systemic embolism occurred in the group receiving dabigatran 150 mg bid (1.11. % per year compared to 1.69% for warfarin and 1.53% for dabigatran 110 mg bid.) Both dabigatran doses were non-inferior to warfarin but only the higher dose was superior. Bleeding rate was highest in the warfarin group (3.36% per year) and lowest in the dabigatran 110 mg bid group (2.71% per year). In the dabigatran 150 mg bid group, the rate was 3.11% per year.

The only significantly increased adverse effect in the dabigatran treated patients was dyspepsia. There were higher rates of myocardial infarction in the dabigatran groups, but not statistically significant. The authors hypothesize that warfarin may have a more protective effect in coronary disease than dabigatran.

There were also higher rates of discontinuation in both dabigatran groups, apparently because of GI symptoms and other unspecified serous adverse events. The authors do not go into any detail to explain the latter.

The quality of warfarin administration was good, as evidenced by a 64% rate of therapeutic INR in the warfarin treated group.

In conclusion: dabigatran appears to be non inferior to warfarin both in terms of efficacy and complications for prophylaxis of stroke and systemic embolism in patients with atrial fibrillation. As part of a new class of oral anticoagulants, dabigatran appears to be headed towards approval for use in a variety of settings. Keep yours eye open for this new drug.

Hemorrhagic complications of influenza

“Then there was the blood, blood pouring from the body. To see blood trickle, and in some cases spurt, from someone’s nose, mouth, even from the ears or around the eyes, had to terrify.

“One of the most striking of the complications was hemorrhage from mucous membranes, especially from the nose, stomach, and intestine. Bleeding from the ears and petechial hemorrhages in the skin also occurred.

“One German investigator recorded ‘hemorrhages occurring in different parts of the interior of the eye’ with great frequency. An American pathologist noted: ‘Fifty cases of subconjunctival hemorrhage were counted. Twelve had a true hemoptysis, bright red blood with no admixture of mucus…Three cases had intestinal hemorrhage…’

These frightening descriptions of “the flu” come from the book The Great Influenza by John M. Barry, an outstanding treatise on the history of the 1918 flu pandemic and the history of American medicine and its response to the medical emergency posed by the pandemic. With current concerns about H1N1 virus, it made me wonder if we might expect similar dramatic clinical manifestations.

But surprisingly (and happily), clinical descriptions of the current H1N1 flu outbreak, which conceivably has the potential to reach pandemic levels, are surprisingly tame with respect to hematologic complications. Both mild leukocytosis and leucopenia have been seen according to reports. A substantial incidence of pulmonary embolism was reported in five of 10 ICU patients hospitalized with H1N1 and ARDS in Michigan and “hypercoagulable state” was reported to be evident in two additional patients. Otherwise, there seems little to report. I will keep watch on this.

Reversal of anti-coagulation--when should you get excited?

Reversal of anti-coagulation--when should you get excited?

Friends and Colleagues:

When I attended the 2008 ASH (American Society of Hematology) annual meeting in San Francisco in

December, I attended a very informative and practical talk by Dr. Mark Crowther of McMaster University on the

topic of anticoagulation reversal.

When you have a patient who is having major bleeding and an INR > 15, it seems pretty clear that immediate

action in the form of factor replacement is needed. What is possibly less well appreciated is that the non-bleeding

patient with prolonged INR of almost any degree rarely if ever should be given infusions of plasma or factor

concentrates.

Dr. Crowther began his talk with fundamental principles: don't lose your cool, stop warfarin, assess the situation, administer antidotes if needed, allow time for anticoagulant drugs to clear.

If the INR is between 5 and 9 and there is no bleeding, the American College of Chest Physicians (ACCP)

recommends holding warfarin and monitoring. If the situation is urgent (i.e. there is high risk of bleeding or

surgery is needed) one can consider giving Vitamin K 1 mg orally (subcutaneous administration is not advised

due to erratic absorption). Dr. Crowther cited four clinical trials that demonstrated that while vitamin K

administration was superior to simple withdrawal of warfarin in reducing the INR, it was not superior in reducing

the incidence of bleeding. Therefore, if you can wait it out, the moral of the story is that simply stopping warfarin and monitoring the INR is an appropriate management strategy, even with an INR as high as 9.

If the INR is higher than 9 and there is insignificant bleeding, higher doses of oral vitamin K (2.5 to 5 mg) and INRmonitoring are suggested.

What if there is major bleeding? The ACCP recommends stopping warfarin, giving IV vitamin K (10 mg, slowly),

giving plasma and/or factor concentrates and repeating IV vitamin K every 12 hours. Dr. Crowther recommended the use of factor concentrates in addition to FFP, since the latter rarely completely corrects the factor deficit,

unless given in unacceptably large volumes. Factor concentrates can provide dramatically fast correction of the INR, in comparison to FFP.

Consensus: Menorrhagia work-up should include consideration of a bleeding disorder

The July 2009 issue of the American Journal of Obstetrics and Gynecology will report the results of a September 2007 consensus conference, with participation by experts in ob-gyn and hematology on the diagnosis and management of reproductive tract bleeding in women with bleeding disorders http://www.ajog.org/article/S0002-9378(09)00410-4/fulltext. The conference was predicated on the belief that the lack of awareness of the potential of bleeding disorders to exacerbate or cause abnormal bleeding leads to underdiagnosis and suboptimal treatment of women with bleeding disorders.
The report brings to light some interesting numbers. For example, the prevalence of menorrhagia in women with von Willebrand Disease (vWD) is between 74 and 92%; conversely, the prevalence of vWD in women with menorrhagia is between 5-24% A study was cited that reported a 33% prevalence of vWD is adolescents with menorrhagia.
The fact that as many as one in every four women with menorrhagia could have a diagnosis of vWD is an impressive one, and surprising. The statistic of one in three for adolescents is even more striking.
The report points out that most women with menorrhagia do not have a bleeding disorder and that even in women with a known bleeding disorder, a gyn evaluation for other causes of bleeding is still mandatory. An underlying bleeding disorder should be considered, however, if any of the following indicators are present:
menorrhagia since menarche
family history of bleeding disorder
personal history of epistaxis lasting more than 10 minutes or necessitating packing or cautery
"notable" bruising without injury
minor wound bleeding, e.g. from trivial cuts, lasting for more than five minutes
bleeding from an oral cavity or GI tract without obvious anatomic lesion
prolonged bleeding after dental extraction
unexpected postsurgical bleeding
hemorrhage from ovarian cysts or corpus luteum, with or without Mittelschmerz
hemorrhage requiring blood transfusion
postpartum hemorrhage, especially when delayed
failure of response to conventional management of menorrhagia
Hematologic evaluation is most sensitive during menstruation, when coagulation factors are at their lowest. However, evaluation should not be delayed until menstruation. Tests can be repeated during menstruation if initial testing is at low normal. Patients should not be removed from contraceptives to permit testing although patients with mild vWD may have normal results when contraceptives are being used. A concluding comment from this article includes the following: "collaboration among obstetrician-gynecologists and hematologists ...could lead to a decrease in the diagnosis of "idiopathic" menorrhagia and allow more effective management of bleeding events." Another conclusion reached is that consideration of bleeding disorders in menorrhagia can lead to improved management of post-partum hemorrhage.

What is the optimum duration of warfarin therapy? Only your ultrasonographer knows for sure...

VTE (venous thromboembolic) disease is a potentially chronic disorder. In one study, the eight year risk of recurrent VTE was 30% in patients with unprovoked events.

The challenge physicians face is to decide who needs longer duration of therapy, and who doesn't, since there is the potential for fatal thrombotic complications in the former group, and serious bleeding complications in the latter. Optimum duration of warfarin therapy has been debated for years, and several different approaches have been taken to make a decision about treatment duration.

In patients whose VTE is provoked by an identifiable and reversible factor (e.g., pregnancy, BCP's, estrogen replacement, recent surgery, recent fracture, long distance air travel, prolonged hospitalization) the duration of therapy can be limited to three to six months. (Recent guidelines published in Chest (http://www.chestjournal.org/content/133/6_suppl) include the option of only three months of therapy for provoked P.E.)

The problem arises in those whose VTE is seemingly unprovoked, or in whom the provocative factor may be a little iffy. These are the patients who deserve a work-up for underlying hypercoagulable state, and who may also benefit from prolonged warfarin therapy, perhaps even lifelong in duration.

In 2003, the NEJM published two articles only several months apart, looking at the results of using prolonged low intensity warfarin therapy (target INR 1.5-2.0), which came to opposing conclusions. In one, the results of the PREVENT trial (http://content.nejm.org/cgi/content/abstract/348/15/1425), a benefit was found in giving prolonged low intensity warfarin even when compared to full intensity anticoagulation. The other trial, named ELATE (http://content.nejm.org/cgi/content/abstract/349/7/631), did not show such a benefit. Thus, the practice of prolonged low intensity warfarin therapy has not really caught on much, although it's something that I do in my practice.

On the other hand, measurement of D-dimer following completion of a specified period of anti-vitamin K therapy, was found to be helpful in predicting whether a person would have recurrence of VTE. The PROLONG study was published in the NEJM in 2006 (http://cme.nejm.org/cgi/content/abstract/nejm;355/17/1780), and this showed that patients with an abnormal D-dimer level one month after the discontinuation of warfarin had a significant incidence of recurrent VTE which can be reduced by the continuation of anticoagulation. Interestingly, the group of patients who had a normal D-dimer had a higher hazard rate of recurrence, compared to the high D-dimer group that received anticoagulation, suggesting that there are yet other factors that may predict long term outcome. Good reviews of the science and clinical application of D-dimer can be found in the journal Blood (http://bloodjournal.hematologylibrary.org/cgi/content/short/113/13/2878) and in the Annals of Internal Medicine (http://www.annals.org/cgi/content/abstract/149/7/481)

Another study (http://jama.ama-assn.org/cgi/content/abstract/296/4/397) looked at thrombin generation time after discontinuation of anticoagulation as a means of determining risk; the patients with lower incidence of recurrent VTE had lower mean peak thrombin generation times. This study, although interesting, has low practical value at the present time.

Most recently, the Annals of Internal Medicine (http://www.annals.org/cgi/content/short/150/9/577) published a provocative article suggesting that the presence of recanalization of veins on follow-up ultrasonography (UTS) is a predictor of long term outcomes for patients with DVT. In this study, 538 patients who had received an initial three months of anticoagulation were randomly assigned to receive either a fixed duration of anticoagulation therapy or flexible duration of therapy whose length was determined by UTS findings at pre-determined intervals. In both the fixed group and the flexible group, the patients were further subdivided into those who had unprovoked DVT or provoked DVT. In the fixed group, the patients with unprovoked DVT received an additional three months of anticoagulation (for a total of six months.) Both groups of patients were subjected to serial UTS up to 21 months if vein recanalization was not observed.

But in the flexible group, the decision to continue anticoagulation was based solely upon serial UTS performed at 3, 9, 15 and 21 months. If the UTS showed resolution of thrombus (defined as vein recanalization), then anticoagulation was discontinued. The mean duration of anticoagulation in the flexible group with provoked DVT was 3.7 months and in unprovoked DVT was 5 months.

There were 78 episodes of recurrent VTE in the 538 patients. The incidence was 17.2% treated in the fixed duration arm, and 11.9% in the flexible duration arm. The cumulative incidence of recurrent VTE was less in the flexible group than in the fixed group. Bleeding complications were higher in the flexible group than in the fixed, but not significantly so.

These results suggested that "assessment of residual thrombosis by ultrasonography improves the identification of patients who are at highest risk for recurrence, and it allowed decisions on anticoagulation that led to a substantial reduction in recurrent events." An interesting note is that prolonging anticoagulation seemed to accelerate vein recanalization, because there was a significantly smaller proportion of patients with persistent vein obstruction in the fixed group than in the flexible group after the last UTS exam in both groups. The authors further conclude that residual thrombosis is a marker of hypercoagulability and that it may predict late cardiovascular death.

This study is a very interesting one (with some limitations) and suggests that follow-up UTS may be a useful adjunct to other clinical evaluations that are used in making the decision whether to continue anticoagulation in patients with DVT. (Note that this study was not about patients with P.E.) In my opinion, this study does suggest that residual thrombi on follow-up UTS should be taken seriously and should prompt continued anticoagulation, and so it does confirm to me the value of my usual practice of obtaining a follow-up UTS after a prescribed course of anticoagulation.

What I was not able to tell from this paper was whether the predictive value of a follow-up ultrasound could be determined early on, or if multiple serial UTS was absolutely required. Clearly this has important economic implications, as well as patient compliance and utilization questions.