| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Radiology, Wake Forest University School of Medicine,
Winston-Salem, NC.
2 Present address: Department of Radiology, University of Utah Health Sciences
Center, 30 North 1900 East #1A071, Salt Lake City, UT 84132-2140.
3 Department of Pathology, Wake Forest University School of Medicine,
Winston-Salem, NC.
4 Department of Urology, Wake Forest University School of Medicine,
Winston-Salem, NC.
5 Division of Urology, University of Utah Hospital and Clinics, Salt Lake City,
UT.
Received March 17, 2006;
accepted after revision August 18, 2006.
Address correspondence to M. E. Heilbrun
(marta.heilbrun{at}hsc.utah.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
3 cm. The purpose of this retrospective study was to evaluate the role
and utility of pretreatment CT-guided biopsy in patients referred for
percutaneous thermal ablation of renal tumors. CONCLUSION. Less than 5% of samples in our study were benign, and 11.8% were nondiagnostic. Biopsy in smaller lesions was less accurate; therefore biopsy is less useful for these renal lesions. Because fine-needle aspiration (FNA) has higher sensitivity than core biopsy, an appropriate algorithm may be to begin with FNA and reserve core biopsy for cases in which an onsite cytotechnologist is unavailable or deems the sample of inadequate cellularity.
Keywords: ablation • biopsy • CT • oncology • percutaneous ablation • renal disease
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
With percutaneous ablative techniques, current practice involves a
preprocedure imaging-guided biopsy to render a specific histopathologic
diagnosis
[6–8].
Reasons for obtaining a biopsy before ablation include lesions treated by
ablation are often 3 cm, a group of masses in which the imaging diagnosis
is less specific [9], the need
to determine appropriate follow-up, avoidance of treatment for benign lesions,
and determining the efficacy of percutaneous ablative techniques as emerging
technologies [7,
10]. The development of better
histologic techniques has increased the specificity of a tissue diagnosis
[11]. The purpose of this
study was to use retrospective data to evaluate the role and utility of
pretreatment CT-guided biopsy in patients referred for percutaneous thermal
ablation of renal tumors diagnosed as renal cell carcinoma (RCC) by imaging at
our institution. The data may then influence future decisions whether biopsy
is necessary to determine optimal treatment algorithms in the setting of
suspected RCC and to further evaluate the diagnostic specificity of imaging
interpreted as "high probability" for RCC.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
Over the observational period, 106 lesions were treated in 103 patients. Of the three patients who underwent multiple treatments, two were treated for multiple lesions, one with a history of von Hippel-Lindau disease and one with multifocal RCC. One treatment was for a recurrence at the site of the original ablation. Thirteen lesions were excluded from analysis, six because the biopsy was performed in a separate session and the results were known before the radiofrequency ablation, five because a biopsy was not performed at all, one because the biopsy and treatment were directed at an area of recurrence in a prior radiofrequency ablation treatment bed, and one because it was characterized solely by sonography. The study cohort therefore is made up of 93 lesions in 91 patients.
Table 1 summarizes the
imaging workup of the 93 lesions. A total of 125 imaging studies were
performed. Most lesions were characterized by CT, which was performed before
and after injection of IV contrast material whenever possible, as determined
by the patient's renal function, age, and comorbidities. Almost half of the
patients underwent MRI, and 32 patients were imaged with both techniques. The
average lesion diameter was 2.9 cm, with a range from 0.7 to 7.8 cm. Two
thirds of the lesions were 3 cm, with 3.2% (3/93) measuring 1 cm,
34.4% (32/93), 1–2 cm, and 31.2% (29/93), 2–3 cm.
|
Tissue was obtained using CT guidance through a 19-gauge needle using a coaxial system (Truguide coaxial biopsy needle, C.R. Bard) with a minimum of two 22-gauge fine-needle aspiration (FNA) biopsies (Chiba biopsy needle, Cook Medical) and a minimum of two 20-gauge core biopsies with an automated biopsy gun (Bard Monopty biopsy instrument, C.R. Bard) immediately before the ablation in 93 cases. For the purposes of this study, the samples described as FNA were those obtained with 22-gauge needles and that underwent a cytologic examination, whereas those categorized as core biopsy samples were obtained with the 20-gauge automated core biopsy gun and were evaluated histologically. The number of samples obtained for both cytologic and histologic evaluation sometimes exceeded these minimums and varied from patient to patient at the discretion of the radiologist performing the procedure. These actual numbers were not systematically recorded in the procedural notes. Figure 1 shows an image from a biopsy procedure. A cytotechnologist was present during the procedure to assess the specimen intended for cytologic examination for adequate cellularity. The samples were then sent to pathology where the same pathologist performed the cytologic and histologic evaluation for each sample. Samples were treated routinely with H and E, and, at the discretion of the interpreting pathologist, additional immunohistochemical studies were performed.
|
Biopsy results were initially classified as nondiagnostic or diagnostic. All biopsy specimens that yielded only benign renal parenchymal cells, necrosis, inflammation, or scant cellular tissue were classified as nondiagnostic. Results classified as diagnostic included RCC, oncocytic neoplasm not otherwise specified (NOS), highly suspicious for RCC, suspicious for RCC, or benign if a specific benign cause such as angiomyolipoma (AML) or oncocytoma was identified on either the FNA or the core biopsy. We asked a pathologist with expertise in urinary tract neoplasms to review all the cases listed as oncocytic neoplasm NOS, highly suspicious, suspicious, nondiagnostic, and benign in an attempt to refine or confirm the diagnosis.
After this review, the diagnostic lesions were further subdivided into benign or malignant categories. Benign lesions included both AMLs and lesions confidently characterized as oncocytoma on the basis of histochemical, immunocytochemical, and ultrastructural studies [11, 14–16]. Malignant lesions included all lesions definitively characterized as RCC. The lesions described in the final pathology report as highly suspicious for RCC or suspicious for RCC were included in the malignant category because the samples contained cells with malignant features but were limited by a low number of collected malignant cells. Also, results described as oncocytic neoplasm NOS that could not be diagnosed as a benign oncocytoma were included in the malignant category after the second review.
Sensitivities of the FNA samples and of the core biopsy samples were calculated using the subset of diagnostic samples in which both an FNA and a core biopsy specimen were obtained. True-positive results were defined as those in which a diagnosis of malignancy was made from a biopsy sample. False-negative results were defined as those biopsies in which a diagnosis of malignancy was made on the alternative mode of biopsy (i.e., FNA was nondiagnostic and core biopsy was diagnostic for RCC). The false-positive rate for the biopsy was defined as zero. Specificities could not be calculated because cases with true-negative results were neither sampled nor treated. These data were also used to calculate the positive predictive value of the pretreatment imaging workup to which each lesion was subjected.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
The biopsy yielded diagnostic tissue in 82 (88.2%) of 93 lesions. A subgroup analysis of the 82 diagnostic specimens was performed to evaluate the efficacy of each biopsy technique. Of these, 77 (73 malignant and four benign) were categorized on the basis of both the FNA and the core biopsy results, four on the basis of FNA alone, and one on core biopsy results alone. FNA produced a diagnostic result in 73 (80.2%) of 91 samples, and the core biopsy was diagnostic in 60 (67.4%) of 89 samples.
Both techniques were performed in 87 lesions, yielding concordant and diagnostic results in both samples in 51 (58.6%) lesions. Eighteen diagnoses were made on the basis of the FNA sample when the core tissue analysis yielded nondiagnostic results. In the four instances in which only FNA sampling was performed, all samples were diagnostic. Eight additional diagnoses were made on the basis of the core tissue sample when the FNA sample was deemed nondiagnostic. In the two instances in which only a core sample was sent for analysis, one sample was diagnostic of RCC and one sample was nondiagnostic. A total of 73 lesions were malignant by a combination of both techniques. Thus, the number of true-positive results by FNA is 65 (73 diagnostic samples minus four results not subjected to core biopsy minus four benign lesions), and the number of true-positive results by core biopsy is 55 (60 diagnostic samples minus one diagnostic sample not confirmed by FNA minus four benign lesions). Thus, the sensitivity of FNA was 89.0%, whereas the sensitivity of core biopsy was lower at 75.3%. Using the data from the diagnostic results, the positive predictive value of the pretreatment imaging workup is 95.1% (true-positive, 78; false-positive, four).
The average size of the lesions with diagnostic results was 2.9 cm, with a range of 0.9–7.8 cm. The average lesion size of the nondiagnostic samples was 2.3 cm with a range of 1.1–4.4 cm. The average size of lesions that resulted in nondiagnostic samples was statistically smaller than the average size of the lesions with diagnostic results (p =0.04).
Diagnostic imaging and the images obtained during the procedure were reviewed retrospectively in all 11 patients in whom the pathologic results were nondiagnostic. Eight were avidly enhancing (> 20 H on CT or > 15% on MRI) solid lesions and three were Bosniak III cystic lesions. Although these tumors were classified in the nondiagnostic category for the purposes of this study, the nondiagnostic results were felt to represent sampling error and the presumptive diagnosis remained malignancy rather than the result of a false-positive imaging diagnosis. All lesions were treated at the time of biopsy without a final pathologic diagnosis.
The biopsy results, including the histologic subtypes, are summarized in Table 3. Sixtyfive lesions were diagnosed as RCC, one deemed highly suspicious for RCC, and five suspicious for RCC, and 10 were diagnosed as oncocytic neoplasm NOS. Of the 82 diagnostic biopsies, the pathologic results yielded a diagnosis of malignant neoplasm in 78 instances. Only one lesion was found unexpectedly to be an AML. CT detected no fat in that 1.7-cm tumor before biopsy and treatment. In addition, three of the lesions originally included in the oncocytic neoplasm NOS were found to have features consistent with a diagnosis of oncocytoma. Two were characterized as being Hale's colloidal iron stain negative, strongly suggesting a diagnosis of oncocytoma. One lesion, despite being Hale's colloidal iron stain negative, was felt to represent an RCC because of the morphology of the cells and the characteristics of other stains. The final diagnosis in the remaining seven oncocytic neoplasms NOS remained inconclusive, with a differential diagnosis of RCC versus oncocytoma, and these were categorized as malignant lesions. Six samples showed features suggestive of RCC but could not be definitely characterized because of scant cellularity and, thus, account for the diagnoses of highly suspicious for RCC and suspicious for RCC on the initial report by the interpreting pathologist. The second reviewing pathologist thought that the likelihood of malignancy was so high that these lesions should be included in the malignant category. Therefore, in the 82 diagnostic samples, 78 (95.1%) were malignant and four (4.9%) were benign.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
In this series, which uses the results of biopsies obtained before an ablation procedure to define the accuracy of the pretreatment diagnosis, the only definitive diagnoses considered were malignant or benign. A tissue diagnosis of atypical or suspicious cells was categorized as malignant when a rereview of each biopsy sample indicated the presence of malignant cells. This group accounted for only 7.2% of the diagnostic results and 6.3% of the entire sample. Likewise, the only results classified as benign were those for which a specific benign diagnosis was made, including AML and oncocytoma. We had a low incidence of definitively benign lesions that would constitute false-positive imaging results. One lesion was a lipid-poor AML and three lesions were oncocytomas. The rate of oncocytomas is lower in our series than in other recent series, which have reported approximately 10% of small renal tumors were oncocytomas [9].
In this study, all alternative diagnoses such as chronic inflammation, fibrosis, and renal parenchymal cells were categorized as nondiagnostic, which we include as a separate category, whereas other studies have listed these lesions in the negative category [20, 21]. Because false-negative results are known to occur, it has been suggested that negative results be viewed with suspicion [10, 22]. For example, Neuzillet and colleagues [8] describe eight lesions that were not categorized on biopsy: five biopsy results were described as fibrosis and three had insufficient tissue. Of those eight lesions, five were resected, and it was found that the biopsy results represented false-negative results, with RCC found at surgical resection. Of the other three tumors, two were stable on surveillance and one lesion was lost to follow-up. One study looked at the results of repeated biopsy of renal tumors with imaging features of RCC that initially yielded nondiagnostic specimens. A large majority of these masses were found to be RCC on subsequent biopsies [23]. These studies emphasize the likelihood that nondiagnostic samples actually represent false-negative biopsies rather than benign histology of the masses.
The difficulty in making a definitive diagnosis on needle biopsy affects the utility and role of biopsy in the routine diagnostic evaluation of the renal mass. A recent study by Dechet et al. [15] compared CT categorization and biopsy categorization using a final whole resected specimen as the gold standard. Both radiologists and pathologists encountered a subset of lesions that could not be categorized as malignant or benign on the basis of either the imaging features or the biopsy sample. The diagnosis was uncertain on imaging more often than the diagnosis was uncertain on needle biopsy. But, more than 70% of the lesions classified as nondiagnostic, either by imaging or needle biopsy, were, in fact, found to be malignant on the basis of the final specimen evaluation. Because of this high degree of inaccuracy, the authors concluded that routine needle biopsies could not be recommended as a means of formulating a treatment plan for small renal tumors.
These findings also have implications for the role of biopsy in the workup of the lesion treated by ablation because the sample sent to pathology may be inadequate for diagnosis, but no additional tissue will be available after treatment. The nondiagnostic component of our sample was 11.8%, the suspicious and highly suspicious component made up another 7.3%, so that nearly 20% of the samples could not be definitively characterized based on the initial analysis of samples obtained at biopsy. But because of the known low negative predicative value of the percutaneous biopsy, it is still recommended that negative biopsies be followed with treatment [21]. If the information provided by a negative biopsy result is not sufficient to exclude a patient from treatment, then the added cost and risk of the biopsy may not be warranted.
The incidence of benign lesions in more recent surgical series describing resected small renal tumors is between 16.5% and 19.2%, raising a question about the accuracy of the imaging diagnosis [8, 20, 24]. It has been shown that as the size of the renal mass decreases, the incidence of benignity increases [2, 7, 9, 25]. Unfortunately, as the size of the mass decreases, the sensitivity of the biopsy decreases [8]. In our series, the size of the lesions biopsied did play a significant role in the ability to obtain sufficient and adequate tissue for a diagnosis. Yet, ablation is most effective for tumors less than 3 cm in size [6, 26]. This would suggest that the precise population most likely to benefit from the ablation may be the population in which the biopsy results are least predictive.
Some have postulated that, as an emerging technology, percutaneous ablative techniques have the potential to be as effective as partial nephrectomy and, in some instances, may become the standard of care for the treatment of small renal masses [3, 6]. An important argument in favor of obtaining a tissue diagnosis is that ablative techniques remain incompletely validated, and the biopsy results are necessary to evaluate the efficacy of the technique [10]. However, retrospective trials looking at diagnostic accuracy are limited by nonstandardized methods.
One recent retrospective study that sought to define the prevalence of benign masses in a population of patients referred for ablation found that 40% of the lesions biopsied were benign [7]. Whereas four of 27 lesions were characterized as lipid-poor AMLs on biopsy, an additional four masses showed only equivocal enhancement (defined by the authors as 10–19 H) before the biopsy. The authors argue that because of the high number of benign lesions encountered on biopsy, the biopsy should be obtained in a separate session from the ablation procedure. This will allow the exclusion of patients with benign lesions from an unnecessary treatment, which would be an important benefit both in terms of risk and total cost to the individual patient from an unnecessary treatment procedure. We concur that this is clearly an important benefit of the biopsy; however, we encountered a significantly smaller number of benign lesions in our cohort, suggesting that by using careful imaging selection criteria, especially when encountering cystic masses or equivocally enhancing masses, benign masses will be encountered at a low rate.
The biopsy procedure itself is quite safe, with a very low rate of complications and no cases of needle tract seeding reported in the literature in the last 10 years (based on a PubMed search of "needle tract seeding AND biopsy AND renal"). Reported biopsy complications include hemorrhage, hematuria, and pain, but the rate is low in recent series, ranging from 0% to 9%. Only minor complications that never require intervention or transfusion are reported in multiple series describing small-gauge biopsy (17–22 gauge) [8, 27, 28]. In our series, we had no complications that we could causally relate specifically to the biopsy component of the procedure because the biopsy was followed immediately by the ablation.
However, the biopsy does add an additional cost to the procedure. In our institution, based on Current Procedural Terminology (CPT) coding, a CT-guided renal mass biopsy will result in a technical charge of $1,070.75 and a professional charge of $621.00 (CPT codes 50200 and 76360). If the biopsy is performed concurrently with the ablation procedure, the multiple surgery rule would apply and the biopsy charge would be reduced by 50%. Previous studies have shown a cost savings of as much as $3,000 per patient for percutaneous biopsy of abdominal masses over open biopsy [29], but we were unable to find a study that looked at the specific added cost of the biopsy to a percutaneous treatment procedure. This type of analysis would be helpful to gain a better understanding of the utility of the biopsy and the role it should play in the percutaneous ablation algorithm.
To determine the true efficacy of percutaneous ablative techniques, prospective trials will be necessary. An important component of the prospective study will be to stratify outcomes based on pathology, including the comparison of results between patients in different study limbs with lesions that are false-positive on imaging. A determination of tumor type and grade is essential in this setting to accurately assess outcomes after radiofrequency ablation when compared with total or partial nephrectomy. However, in a retrospective study, this information is less reliable because of the considerable differences in techniques, pathologic interpretations, and sensitivities of the various pathologic staining methods.
Traditionally, tumor stage, nuclear grade, and tumor cell type are the main prognostic clinical parameters in the setting of RCC. The difference between a T1 lesion and a T3 lesion may be as small as microvascular involvement and microscopic extension beyond the renal capsule. A recent study found that the presence of microscopic invasion beyond the renal capsule, despite being occult on imaging, did not significantly alter the prognosis for lesions considered stage T1 by CT [30]. If these small lesions usually behave like a T1 lesion regardless of the actual pathologic classification, the tissue diagnosis may have little bearing on the outcome for an individual patient. The biopsy adds cost and incremental risk to the ablation procedure but does not change the prognosis or influence the outcome.
An unexpected finding in our study was that FNA was more often diagnostic than the core biopsy sample despite the additional architectural information available with histology. This may be related to technique because FNA involved obtaining multiple passes with each needle throughout the lesion, whereas each core biopsy sampled only one area. The 20-gauge size of the core biopsy samples may also have decreased the efficacy of this biopsy technique. In addition, because of the size of the lesion and because we performed the FNA first, the tissue obtained during a core biopsy may already have been damaged and therefore may be more difficult to interpret. These results suggest that it may be possible to begin with the FNA and ask the onsite cytotechnologist to examine the specimens and only proceed to a core sample if the specimen is thought to contain inadequate cells for characterization. This approach is also that used by Silverman et al. [10] who have considerable experience in both percutaneous biopsies and percutaneous ablation of renal masses.
Limitations of this study include the retrospective nature of the study and the lack of a standardized pathologic regimen for classifying the biopsy samples. Specifically, not all of the samples were treated with additional staining techniques to refine the diagnosis of oncocytic neoplasm. And the significance of additional pathologic evaluation is not universally agreed on by pathologists [11, 31]. In addition, in our categorization of the pathology results, we considered only three possible results: benign, malignant, or nondiagnostic. We made the assumption that nondiagnostic biopsy results represented false-negative findings. This aggressive categorization may result in placing benign lesions into a malignant category, which reduces the utility of the biopsy and subsequently overstates the benefits of the ablation procedure.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. J. Stanley Our Practice of Radiology: Reflections on its Growth and Stature Am. J. Roentgenol., June 1, 2007; 188(6): 1439 - 1439. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
