| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Diagnostic Radiology, University of Maryland School of Medicine,
22 S. Greene St., Baltimore, MD 21201.
2
Greenebaum Cancer Center, University of Maryland, Baltimore, MD 21201.
Received February 26, 1999;
accepted after revision June 22, 1999.
Address correspondence to W. A. Berg.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Experience with 501 14-gauge and 179 11-gauge core needle breast biopsies was recorded prospectively. We performed 18 (3%) of 680 procedures on patients undergoing anticoagulation therapy. Eleven biopsies were performed in eight women treated with warfarin, one in a patient treated with heparin, and six in six women treated with aspirin. Ten biopsies were performed with sonographic guidance and a 14-gauge automated biopsy gun. Eight biopsies were performed with an 11-gauge vacuum-assisted probe and stereotactic guidance.
RESULTS. Hematomas occurred in three (38%) of eight anticoagulated patients undergoing 11-gauge vacuum-assisted stereotactic biopsy, measuring 13 mm, 17 mm, and 40 mm. In patients not known to be anticoagulated, small hematomas were mammographically evident in 77 (45%) of 171 biopsies performed with an 11-gauge probe, averaging 8 mm (range, 2-17 mm); only 13 (8%) of 171 had hematomas larger than 10 mm. Small hematomas, averaging 6 mm (range, 2-12 mm) were evident mammographically in 45 (45%) of 100 stereotactic biopsies using a 14-gauge automated biopsy gun; three (3%) experienced hematomas larger than 10 mm. A 10-mm hematoma was evident during one (10%) of the 10 biopsies performed in patients undergoing anticoagulation therapy under sonographic guidance using a 14-gauge automated biopsy gun; small hematomas (mean, 10 mm) were noted during seven (2%) of 391 corresponding sonographically guided biopsies in patients not undergoing anticoagulation therapy.
CONCLUSION. No patients undergoing anticoagulation therapy experienced clinically important complications; preliminary results suggest rates of hematoma formation similar to those in the control group. As such, discontinuing anticoagulation medication before core needle breast biopsy may be unnecessary when the need for biopsy is urgent.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Postprocedural digital spot mammographic films were obtained during all stereotactic biopsies and hematoma size was measured. For sonographically guided biopsies, hematomas evident sonographically at the time of procedure were documented. When results were telephoned to the patient 18-28 hr after biopsy, all patients were asked if they had experienced any problems such as bleeding or lump at the biopsy site. Three patients telephoned us within 2 weeks of the biopsy to report a lump (one taking warfarin and two in the group without anticoagulation); repeated sonography was then performed.
We biopsied eight women undergoing anticoagulation with warfarin (three had two areas sampled), one with heparin, and six with aspirin. Ten suspicious masses were biopsied under sonographic guidance (six undergoing therapy with warfarin, one with heparin, and three with aspirin) and eight indeterminate clusters of calcifications were biopsied using the 11-gauge vacuum-assisted probe and stereotactic guidance (five undergoing anticoagulation therapy with warfarin and three with aspirin). An average of 5.3 passes were made for sonographically guided biopsies (range, three to seven passes) and 16 passes for the stereotactic biopsies (range, 10-30 passes).
For patients undergoing anticoagulation therapy with heparin or warfarin, recent prothrombin time (PT), partial thromboplastin time (PTT), and International Normalized Ratio (ratio of patient's PT and the mean PT for a group of healthy individuals) were reviewed. A bleeding time is not routinely obtained for patients undergoing anticoagulation therapy with aspirin, because these patients are instructed to not take aspirin for 7 days before a procedure.
Standard deviation was calculated for the rates of hematoma formation for patients with and without anticoagulation therapy using a binomial formula. Fisher's exact test was used to calculate p values.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
The six patients undergoing anticoagulation therapy with aspirin and one with warfarin did not have clotting studies performed. As stated, patients undergoing anticoagulation therapy with aspirin are not routinely evaluated with a bleeding time before biopsy. Of the remaining eight patients, seven had PT values, PTT values, or both obtained within 10 days before biopsy. One patient had PT and PTT values assessed 2 months before biopsy. The normal range for PT is 10.9-13.1 sec, for PTT 22.0-36.0 sec, and for International Normalized Ratio 1.0. Six biopsies were performed with elevated values for both PT and PTT, with values ranging from 14.0 to 21.1 sec (mean, 17.4) and 37.0 to 55.0 sec (mean, 43.2), respectively. Four were performed with elevated PT values ranging from 15.0 to 26.2 sec (mean, 18.2) and normal PTT. One biopsy was performed where only the PTT value was elevated to 65 sec. The International Normalized Ratio was elevated in seven women, ranging from 1.3 to 4.9 (mean, 2.2).
Bleeding During Biopsy
No patients in either the group undergoing anticoagulation or the group
without anticoagulation had clinically significant bleeding or hematoma
formation (requiring a hospital admission or surgical drainage). For all
biopsies performed on patients undergoing anticoagulation therapy, the
patients tolerated the procedures well. During the biopsy, 44 (7%) of the 662
patients in the control group had minimal to moderate bleeding during the
procedure or immediately after probe withdrawal, compared with two (11%) of
the 18 patients in the anticoagulated group. Compression was routinely
maintained for 15 min in anticoagulated patients, and in one biopsy (5.5%) it
took the full 15 min for bleeding to stop. In patients not undergoing
anticoagulation therapy, 24 (14%) of 171 biopsies performed with the 11-gauge
probe were complicated by moderate bleeding requiring 10-15 min of manual
compression. Of the 100 stereotactic biopsies performed with a 14-gauge
automated biopsy gun, four (4%) were complicated by moderate bleeding during
the procedure and 16 (4%) of 391 had moderate bleeding during 14-gauge
sonographically guided biopsy that required 15 min of compression.
Hematomas Depicted Immediately After Biopsy
Of the 10 sonographically guided biopsies in patients undergoing
anticoagulation therapy, one 10-mm hematoma occurred after the biopsy
(Table 1). In three (38%) of
the eight 11-gauge stereotactic biopsies a hematoma was noted
mammographically: one measured 13 mm, one 17 mm, and one 20 mm
(Fig. 2A,
Fig. 2B).
|
|
|
In patients not known to be undergoing anticoagulation therapy, small hematomas were mammographically evident in 77 (45%) of 171 biopsies performed with an 11-gauge probe, with average size of 8 mm (range, 2-17 mm); only 13 (8%) of 171 were complicated by hematomas larger than 10 mm. Moderate bleeding requiring 15 min of compression complicated another seven (4%) of 171. Mammographically evident hematomas averaging 6 mm (range, 2-12 mm) were documented in 45 (45%) of 100 stereotactic biopsies performed with the 14-gauge automated biopsy gun; only three (3%) of 100 biopsies were complicated by hematomas larger than 10 mm. Moderate bleeding requiring 15 min of compression was seen in another six (6%). Small hematomas (mean, 10 mm; range, 7-20 mm) were noted during six (2%) of 391 sonographically guided biopsies in patients not undergoing anticoagulation therapy.
Overall, the rate of imaging-depicted hematomas was similar in both anticoagulated and control groups, but the small sample size results in a large standard deviation for the measurements in the anticoagulated group. Specifically, the rate of hematomas in anticoagulated patients undergoing stereotactic biopsy with 11-gauge probes was 0.38 ± 0.17. This rate is not different from that which we observed for the control group, 0.45 ± 0.03.
For patients undergoing sonographically guided biopsy with a 14-gauge automated biopsy gun, the rate was 0.10 ± 0.09 for anticoagulated patients compared with 0.018 ± 0.007 for the control group. Again, the one hematoma we did see measured 10 mm and was of no clinical significance. No relationship was seen between individual elevated clotting parameters and hematoma formation, although the number of patients was small.
When the rate of hematoma formation is categorized by the type of anticoagulation therapy, a hematoma was seen in two (18%) of 11 biopsies of eight patients undergoing anticoagulation therapy with warfarin, two (33%) of six patients taking aspirin, and no patients taking heparin. In the control group, hematomas were identified in 129 (19%) of 662 patients immediately after biopsy. A trend toward a higher rate of hematoma formation was seen in patients with aspirin than in the control group, but it was not significant (p = 0.4). When the subgroup undergoing sonographically guided core biopsy is analyzed, one (33%) of three patients undergoing aspirin therapy developed a hematoma compared with seven (2%) of 391 in the control group (p = 0.001); with the small number of patients biopsied while undergoing aspirin therapy, such a result must be considered preliminary.
Delayed Hematomas
One of the women biopsied stereotactically with an 11-gauge probe while
undergoing warfarin therapy experienced moderate bleeding after the second
specimen was obtained. Fibrocystic changes with microcalcifications were seen,
as was intimal proliferation due to diabetes in an artery in one of the core
specimens that may have prevented vasospasm. A 20-mm hematoma was evident
immediately after the procedure, and 1 week later the patient was seen because
of erythema and enlarging mass. A 40-mm collection was evident, which could
not be aspirated with an 18-gauge needle under sonographic guidance, and a
presumptive diagnosis of cellulitis and hematoma was made. Cephalexin
antibiotics were prescribed, which the patient did not take, and the symptoms
resolved within 2 weeks without treatment. Interestingly, she presented with
inflammatory breast carcinoma 16 months later throughout the same breast.
One patient in the control group reported a lump 3 days after 11-gauge biopsy after going swimming; a 30-mm hematoma was seen sonographically and resolved after 1 week of warm compresses. One other patient not treated with anticoagulation reported a lump at the biopsy site 1 week after sonographically guided biopsy. Sonography showed a 15-mm hematoma that resolved after 1 week of warm compresses.
Histopathology
In this series of biopsied lesions, 156 (23%) of 680 proved malignant.
High-risk lesions (atypical ductal hyperplasia, lobular neoplasia) were found
in another 18 (3%) of 680 core biopsies. At excision, cancer was proven in
four (44%) of nine lesions yielding only atypical ductal hyperplasia on core
biopsy and in one (11%) of nine lesions yielding only lobular neoplasia on
core biopsy. Eleven (61%) of 18 biopsies in patients undergoing
anticoagulation therapy yielded malignant results: five infiltrating ductal
carcinomas, two infiltrating and intraductal carcinomas, two poorly
differentiated carcinomas, three ductal carcinomas in situ, one duct ectasia
with microcalcifications, one fibrocystic changes with microcalcifications,
two fibrosis with microcalcifications, one fibroadenoma, and one granulation
tissue. Overall, 33 (66%) of 50 malignancies biopsied with stereotactic
guidance were ductal carcinoma in situ and 17 (34%) were infiltrating. Of the
50 malignancies, 42 (84%) manifested as clustered microcalcifications and
eight (16%) as masses. By comparison, eight (8%) of 106 malignant masses
diagnosed by sonographically guided biopsy were ductal carcinoma in situ and
98 (92%) were infiltrating.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Hematoma formation has been reported after core needle breast biopsy but rarely requires intervention. Parker et al. [3] described a multiinstitutional study of 14-gauge automated core breast biopsy in which the researchers found a rate of 0.2% (six of 3765 cases) for significant complications that required surgical or medical intervention. These complications included three hematomas that were drained and three infections requiring drainage, antibiotic therapy, or both. In another study, Parker and Klaus [8] reported complication rates of 1.1% and 1.4% when using 14-gauge and 11-gauge vacuum-assisted directional probes, respectively. Three complications occurred after biopsy with the 14-gauge probe and included one painful hematoma and two cases of severe pain either during the biopsy or for several days thereafter [8]. One nonpainful hematoma occurred in a patient biopsied with the 11-gauge probe 2 days after biopsy while exercising [8]. Burbank [9] studied a series of 269 biopsies performed with the 14-gauge and 71 with the 11-gauge vacuum-assisted directional probes and found significant ecchymoses larger than 5 cm in 4.3% of patients, significant hematomas in less than 1%, and discomfort after the biopsy in 5.4% of patients undergoing breast biopsy.
Harlow et al. [10] evaluated sonographic changes in the breast 2-9 days after core biopsy and showed that fluid collections were found in 23% of patients; six (19%) of these 31 patients had sonographic changes suggesting hematoma. In standard practice, our patients are not reexamined with sonography after biopsy and our results likely underestimate the true incidence of hematoma formation. It is reasonable, however, to conclude that no clinically significant complications were seen.
Liberman et al. [11] evaluated mammographic findings after stereotactic 14-gauge biopsy and found evidence of a hematoma in 60% of patients, with a mean size of 2.4 cm in those that were measurable, similar to our results. Clinically evident hematomas were noted in four lesions (4%), three of which required termination of the biopsy because of bleeding. In the fourth case, the patient developed an expanding hematoma. None of the clinical hematomas required intervention, and all resolved with ice and compression.
The most significant risk related to bleeding and core needle breast biopsy is likely related to obscuring the lesion. We have noted that lesions that are palpable but small can be more difficult for the surgeon to identify reliably in the setting of ecchymosis and edema after core biopsy. Indeed, one case report [12] describes the complication of a large 14 x 12 cm hematoma immediately after a 14-gauge core needle breast biopsy that obscured the lesion, making subsequent needle localization difficult. On hematologic analysis, coagulation factor XI deficiency was discovered.
Patients at risk for bleeding include those with coagulation factor VIII or
XI deficiency, disseminated intravascular coagulation, uremia, liver failure,
myeloproliferative disorders, monoclonal gammopathy,
2-antiplasmin deficiency, and lupus anticoagulant
[13]. Under ideal
circumstances, patients who are to undergo core needle breast biopsy should
have coagulation values that are within normal limits. We do not typically
obtain coagulation values before biopsy, unless the patient is known to be
undergoing anticoagulation therapy or has any of the aforementioned risk
factors. Patients undergoing anticoagulation therapy with aspirin in this
study did not have bleeding times obtained before biopsy, because these
patients are instructed to not take aspirin for 7 days before the biopsy.
Occasionally patient circumstances may warrant proceeding with the biopsy. Aspirin can generally be stopped for 7 days before biopsy and held for 3 days after without concern for thrombotic events unless the patient refuses to comply. One of the patients undergoing aspirin therapy who also underwent biopsy had three lesions highly suspicious for recurrent cancer 18 months after lumpectomy. She had refused radiation therapy after the initial surgery. When she came to our institution for examination, we performed core biopsy of the largest mass without complication and confirmed recurrent tumor.
If discontinuing anticoagulation with warfarin is deemed undesirable, standard protocol for any biopsy at our facility requires hospitalization and IV heparin, which can then be stopped 4 hr before biopsy and reinstituted later the same day. Such an approach adds delays and cost to diagnosis. If core needle breast biopsy can reasonably be performed while patients are undergoing anticoagulation therapy with warfarin, it is desirable.
If bleeding is encountered during performance of a core needle breast biopsy, manual compression is applied. In our experience, compression is more readily achieved in sonographically guided procedures (unless the breast is very large), because patient position and the probe may impede access to the area during stereotactic biopsies. Further, images are not routinely obtained during stereotactic biopsy until all samples have been obtained and the operator may be less aware that a hematoma is forming than during sonographically guided biopsies where real-time imaging is performed throughout the biopsy and images are documented before and after firing for each sample. In our experience, both stereotactic and sonographically guided biopsies could be performed safely in patients undergoing anticoagulation therapy.
The high rate of malignancy in patients biopsied while undergoing anticoagulation therapy in our series reflects a selection bias. Biopsy of low-suspicion lesions was deferred until anticoagulation medication could be stopped whenever possible. Although intraoperative frozen section is another option in these patients, tissue is irretrievably sacrificed (which presents problems for small lesions), and errors in interpretation are more likely due to artifacts. Perhaps most important, participation of the patient in planning definitive surgery, to include sentinel node or other axillary nodal sampling, is facilitated by preoperative core needle biopsy diagnosis.
We deem our results preliminary because of the small number of patients undergoing anticoagulation therapy who have had core needle breast biopsies at our institution.
In our preliminary experience with core needle breast biopsy in patients undergoing anticoagulation therapy, we found that 14-gauge automated core biopsy and 11-gauge vacuum-assisted directional biopsy can be performed safely. We have seen no increased risk of hematoma formation compared with that of the general population. Rarely, a delayed hematoma can be observed. Whenever reasonable, anticoagulation therapy should be discontinued before biopsy. We still recommend discontinuing aspirin for 7 days before biopsy. If, however, biopsy is urgently needed and the patient would be placed at significant risk by discontinuing anticoagulation therapy, our experience suggests that cautiously proceeding may be reasonable. Unless the breast is very large, effective compression can generally be administered should bleeding be observed.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Somerville, P. J. Seifert, S. V. Destounis, P. F. Murphy, and W. Young Anticoagulation and Bleeding Risk After Core Needle Biopsy Am. J. Roentgenol., October 1, 2008; 191(4): 1194 - 1197. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. M. Lomoschitz, T. H. Helbich, M. Rudas, G. Pfarl, K. F. Linnau, A. Stadler, and R. J. Jackman Stereotactic 11-gauge Vacuum-assisted Breast Biopsy: Influence of Number of Specimens on Diagnostic Accuracy Radiology, September 1, 2004; 232(3): 897 - 903. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
