- Email: [email protected]
Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
original articles
Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient E. PRATSa, P. RAZOLAa, J.M. SAINZb, L. TARDÍNa, A. ANDRÉSa, M.D. ABÓSa, F. GARCÍAa, J. FERNANDEZb, L. VILLAVIEJAb AND J. BANZOa Departments of aNuclear Medicine and bRadiodiagnosis. Hospital Clínico Universitario Lozano Blesa. Zaragoza. Spain.
Abstract.—Objective. Establish the usefulness of scintimammography (SM) in day-to-day clinical practice. Material and methods. We have evaluated 308 SM consecutively performed in our Department. The diagnosis was established by way of biopsy or clinical and mammography follow-up (minimum 18 months). In all cases we evaluated the kind of lesions (palpable or non-palpable) and their degree of suspicion of malignancy in the mammography (BIRADS). We determined the number of fine-needle aspiration cytology (FNAC) or biopsies performed in dependence of kind of lesions, mammography and SM results. Results. Final diagnosis was cancer in 18 % of the patients, and 71 % of the lesions were non palpable. Negative predictive value (NPV) of SM was 96 %. Pathological diagnosis was performed in 100 % of lesions with SM+ and 16 % of lesions with SM–. According to the kind of lesion, FNAC and/or biopsy was performed in 62 % of palpable lesions and in 20 % of non-palpable lesions; and according to the mammography in 21 % of lesions BIRADS I-II, 14 % in BIRADS III, 70 % in BIRADS IV and 100 % in BIRADS V. The number of FNAC and/or biopsies performed is significantly higher in all cases when SM is positive (p > 0.001), excepting in BIRADS V lesions. Conclusions. In day-to-day clinical practice SM is generally performed in low prevalence of breast cancer population, in the evaluation of non-palpable lesions with a low suspicion of malignancy. SM has a high NPV, increase the diagnostic accuracy of the mammography and has repercussion on patient management, except in BIRADS V lesions.
Material y método. Hemos revisado 308 GM realizadas en nuestro servicio. El diagnóstico se estableció mediante citología y/o biopsia o por seguimiento clínico/mamográfico superior a 18 meses. En todos los casos se valoró el tipo de lesión (palpable o no) y su sospecha de malignidad en mamografía (BIRADS). Se determinó el número de punciones aspiraciones con aguja fina (PAAF) y/o biopsias realizadas en dependencia del resultado de la GM, tipo de lesión y mamografía. Resultados. El diagnóstico final fue de cáncer en el 18 % de las pacientes, siendo no palpables el 71 % de las lesiones. El valor predictivo negativo (VPN) de la GM fue del 96 %. Se realizó diagnóstico anatomopatológico en el 100 % de lesiones con GM positiva y en el 16 % de lesiones con GM negativa. Dependiendo del tipo de lesión, se practicó PAAF y/o biopsia en el 62 % de lesiones palpables y en el 20 % de las no palpables; según la mamografía, en el 21 % de lesiones BIRADS I-II, 14 % BIRADS III, 70 % BIRADS IV y 100 % BIRADS V. El número de estudios anatomopatológicos practicados fue significativamente mayor en todos los grupos con GM positiva (p > 0,001), excepto en BIRADS V. Conclusiones. En la práctica clínica, la GM se realiza fundamentalmente en poblaciones con baja prevalencia de cáncer, en lesiones no palpables de baja sospecha de malignidad. La GM tiene un alto VPN, incrementa la seguridad diagnóstica de la mamografía y repercute en el manejo de las pacientes, excepto en lesiones BIRADS V. PALABRAS CLAVE: gammagrafía de mama, 99mTc-MIBI, cáncer de mama.
KEY WORDS: scintimammography, 99mTc-MIBI, breast cancer. GAMMAGRAFÍA DE MAMA CON 99MTC-MIBI EN LA PRÁCTICA CLÍNICA. REPERCUSIÓN EN EL MANEJO CLÍNICO DEL PACIENTE Resumen.—Objetivo. Valorar el papel de la gammagrafía de mama (GM) en la práctica clínica. Correspondence: E. PRATS. Servicio de Medicina Nuclear. Hospital Clínico Universitario Lozano Blesa. Avda. San Juan Bosco, 15. 50009 Zaragoza. Spain. E-mail: [email protected] Received: 18-10-06 Accepted: 24-01-07
Mammography is the most useful procedure in the diagnosis of breast cancer, both in the screening examination and when faced with clinical suspicion of the disease.1 However, it has some limitations. Its sensitivity may be reduced in certain situations, such as dense breasts or presence of scar tissue, the cancer rate of the interval described varying between 3 % and 20 %.1-4 Furthermore, mammography is a relatively unspecific diagnostic technique which often does not distinguish malignant lesions from benign pathology.5 Its positive predictive value (PPV) varies between 20 % and 40 %, which leads to an elevated number of biopsies being performed on benign lesions.6,7
Rev Esp Med Nucl. 2007;26(3):153-9
153
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
99m
Tc-MIBI scintimammography (SM) was introduced into clinical practice in 1994 after studies by Waxman et al.8 and Khalkhali et al.9,10 Since then, many papers analyzing the methodology, results or indications of this technique have been published.11-22 However, there is still little information on the real role of this technique in daily clinical practice, evaluating the type of lesions studied, their results or their repercussion in the clinical management of the patient. SM has been used in our hospital as a diagnostic technique complementary to mammography since the end of the last decade. In this setting, the objectives that we proposed in this paper, on assessment of a large series of patients studied in the same centre following a routine diagnostic methodology, were the following: 1) To analyze the study patient sample, behavior of the lesions in the physical examination and mammography and the SM results, based on global results and depending on the different types of lesions, and 2) To assess the repercussion of SM on the clinical management of patients.
MATERIAL AND METHODS
Patients and study design
We retrospectively analyzed 350 SMs performed consecutively in our Nuclear Medicine Department. The patients were referred from different departments of the hospital or from attached specialty centers. Study inclusion criteria were: 1) Availability of the patient’s physical examination data, assessing whether the lesion was palpable or not, and a recent mammography and SM; and 2) Establishment of a definitive diagnosis by cytology and/or biopsy. When no anatomical-pathological study was available, the diagnosis of “benignancy” was made by clinical/mammographic follow-up for a period of at least 18 months. In accordance with these criteria, the total number of SMs included was 308. All patients were women with a mean age of 54 years (range 26-84 years). Mammography
Mammography was performed with cranial-caudal and mediolateral oblique views in all cases, taking images in the lateral view, magnified or compressed when the radiologist considered it necessary. In all the mammographies, the presentation patterns and 154
diagnostic category were assessed in accordance with the American College of Radiology Breast Imaging Reporting and Data System (BIRADS). The diagnostic categories of BIRADS are: I (normal), II (benign), III (probably benign), IV (suspicion) and V (high suspicion of malignancy).23 Scintimammography
The SM was carried out 5 minutes after the intravenous injection of 740 MBq of 99mTc-MIBI (hexakis 2 methoxyisobutyl isonitrile, Cardiolite, Bristol Mayers) in a dorsal vein of the foot. The radiotracer was administered through a butterfly needle and washed with 5-10 cc of physiological serum. SMs were acquired using Elscint SP6, Elscint Helix SPX and Millenium VG General Electric gammacameras. Lateral images of both breasts in anterior and prone decubitus positions were obtained using a specially designed table with lateral openings which allow the breast to fall freely, thus obtaining the maximum separation between deep breast tissue and organs with physiological uptake of 99mTc-MIBI, such as the liver or myocardium. The acquisition time per image was 10 minutes using a high-resolution collimator and 256 × 256 matrix. All focal or multifocal uptake of the radiotracer that was greater than the breast background was considered positive, regardless of the degree of uptake. Diffuse uptakes were not considered suggestive of malignancy. All SMs were reviewed independently by two Nuclear Medicine specialists. When they disagreed, the opinion of a third specialist was sought. Statistical analysis
We used the χ2 test with Yate’s correction to study if there were statistically significant differences between the number of biopsies and/or FNAC performed according to the result of the SM, both in the global sample and depending on the type of lesion studied (palpable or not) and its diagnostic category on mammography (BIRADS). We considered values of p < 0.01 as significant.
RESULTS
Diagnoses were established with anatomicalpathological criteria (cytology and/or biopsy) in
Rev Esp Med Nucl. 2007;26(3):153-9
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
99 patients (32 %). Presence of breast cancer was demonstrated in 56 patients (18 % of the total): 44 invasive ductal carcinomas (IDC), 9 invasive lobular carcinomas (ILC) and 3 in situ ductal carcinomas (DC in situ); no malignant pathology was found in the remaining 43 (14 %) patients: 24 fibrocystic breast disease (FBD), 7 fibroadenomas, 5 post-surgical scars, 2 fat necrosis, 1 scar associated with an inflammatory process, 1 hamartoma, 1 post-traumatic necrosis and 2 with no detectable breast pathology. Malignancy was excluded in the remaining 209 patients (68 %) as no breast pathology was demonstrated during the clinical/ mammographic follow-up of at least 18 months. The number of benign lesions to be studied was therefore 252 (82 %). Eighty-nine lesions
were detected with palpation (29 %) while 219 lesions were not palpable (71 %). The mammographies were assessed in accordance with the following BIRADS categories: I in 18 cases (6 %), II in 28 (9 %), III in 180 (58.5 %), IV in 63 (20.5 %) and V in 19 (6 %). SM was positive in 60 lesions (48 true positive [TP], 12 false positive [FP]) and negative in 248 (240 true negative [TN], 8 false negative [FN]) (Figs. 1-3). Thus, the global results obtained were: sensitivity 85.7 %, specificity 95.2 %, PPV 80 %, negative predictive value (NPV) 96.8 % and diagnostic accuracy 93.5 %. Six of the FN results occurred in lesions of less than 1 cm in diameter (4 IDC, 1 ILC and 1 DC in situ), while the remaining 2 neoplasia
A
B
FIG. 1.—Patient with palpable nodule in inferior-external right breast quadrant. Mammography (A): Extremely dense breast without obvious nodules. Scintimammography (B): Uptake focus coinciding with palpable lesion. Definitive diagnosis: Invasive ductal carcinoma (biopsy).
A
B
FIG. 2.—Fifty-five year-old woman without palpable breast lesion. Mammography (A): Asymmetric density increase in left breast retroareolar region (BIRADS III) with bilateral nodules of benign aspect. Scintimammography (B): Normal. Clinical/radiological follow-up: > 18 months without showing breast pathology. Rev Esp Med Nucl. 2007;26(3):153-9
155
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
A
B
FIG. 3.—Patient operated on for invasive ductal carcinoma 2 years ago. Mammography (A): Post-surgical scar in upper right breast quadrant (BIRADS III). Scintimammography (B): Normal. Clinical/radiological follow-up > 18 months, without showing breast pathology.
not detected in the SM were 2 IDC of between 1 and 2 cm. The anatomical-pathological results corresponding to the FP results were: 7 FBD, 1 fibroadenoma, 1 hamartoma, 1 fat necrosis, 1 post-traumatic necrosis and 1 post-surgical scar with associated inflammatory process. In the 89 palpable lesions (Fig. 1), SM was positive in 42.5 % of the patients (33 TP, 5 FP), and negative in 57.5 % (50 TN, 1 FN), while in the 219 nonpalpable lesions (Figs. 2 and 3), SM was positive in 10 % (15 TP, 7 FP) and negative in 90 % (190 TN, 7 FN). Global results and the results of palpable and nonpalpable lesions are shown in Table 1. If we analyze the size of the breast cancers included in our study, we find 12 T1 a-b (< 1 cm), 24 T1c (1-2 cm) and 20 T2 (> 2 cm). SM detected 6 T1a-b (sensitivity 50 %), 22 T1c (sensitivity 91.7 %) and 20 T2 (sensitivity 100 %). The number of breast cancers found in each BIRADS class was: 2 in BIRADS I (9 %), 1 in BIRADS II (4 %), 9 in BIRADS III (5 %), 25 in BIRADS IV (40 %) and 19 in BIRADS V (100 %). Correlation of SM results with the BIRADS classes shows that the SM was positive in 13 % (3 TP, 3 FP) and negative in 87 % (40 TN) of the 46 BIRADS I or II lesions. Among the 180 BIRADS III lesions (Figs. 2 and 3), the SM was positive in 6.6 % (8 TP, 4 FP) and negative in 93.4 % (167 TN, 1 FN). Among the 63 BIRADS IV, the SM was positive in 38 % 156
(19 TP, 5 FP) and negative in 62 % (33 TN, 6 FN). Finally, among the 19 BIRADS V lesions, the SM was positive in 94.7 % (18 TP) and negative 3.3 % (1 FN). The global results of the SM and the relationship with each type of mammographic lesions according to the BIRADS criteria are shown in Table 2. The SM allowed detection of 11 breast cancers in which the mammography was not suspicious of malignancy (3 BIRADS I/II and 8 BIRADS III). The pathological studies conducted according to SM results for the global sample and according to lesion type (palpable or nonpalpable) and its diagnostic classification on mammography are described in Table 3. As can be seen in the Table, the number of biopsies and/or cytology performed is significantly greater in all the groups when the SM is positive (p < 0.00001 in the global sample, palpable lesions, nonpalpable lesions and lesions with BIRADS I, II and II mammography, and p = 0.00014 in BIRADS IV), except in BIRADS V.
DISCUSSION
Some data of interest regarding the type of lesions studied, their presentation on mammography, clinical results of the SM in different types of lesions and how the SM influences the clinical management of patients can be obtained by analyzing our results. In first place, we can note the high proportion of benign lesions studied with SM. A cancer diagnosis was established in only 56 (18 %) of all the lesions while 252 (82 %) were benign. The percentage of malignant lesions was less than that generally found in the literature. In a meta-analysis and review of the SM results that included 64 articles published up to 1999,24 56 % of the lesions were diagnosed as cancer. Equally, if we analyze multicentre studies, we see that the proportion of malignant lesions varied between 40 % and 84 %.18,25 In a recent study published by Sampalis et al26 which included 1,243 patients, only 16 % had malignant pathology, which is more in accordance with our study. If we consider that SM is a technique complementary to mammography, its use outside of research studies or to assess the response to chemotherapy would be little justified in population groups with high percentages of breast cancer. In most publications on SM, there is a clear predominance of examinations performed in palpable lesions. In the first studies, nonpalpable lesions varied
Rev Esp Med Nucl. 2007;26(3):153-9
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
Table 1 GLOBAL RESULTS OF THE SCINTIMAMMOGRAPHY IN PALPABLE AND NONPALPABLE LESIONS Lesion type
Sensitivity (%)
Specificity(%)
Accuracy(%)
PPV(%)
NPV(%)
Palpable (n = 89) Nonpalpable (n = 219) Total (n = 308)
97 68 85
94 96 95
93 93 93
91 68 80
98 96 96
NPV: negative predictive value; PPV: positive predictive value.
Table 2 RESULTS OF THE SCINTIMAMMOGRAPHY ACCORDING TO THE BIRADS OF THE LESION BIRADS
Sensitivity (%)
Specificity (%)
Accuracy (%)
PPV (%)
NPV (%)
I-II (n = 46) III (n = 180) IV (n = 63) V (n = 19) Total
100 89 76 94 85
93 97 87 – 95
93 97 82 94 93
50 66 79 100 80
100 99 85 –0 96
BIRADS: Breast Imaging Reporting and Data System; NPV: negative predictive value; PPV: positive predictive value.
Table 3 PATHOLOGICAL STUDIES PERFORMED ACCORDING TO SCINTIMAMMOGRAPHY RESULTS
Global SM+ SM– p
Global (%)
Palpable lesions (%)
Nonpalpable lesions (%)
BIRADS I-II (%)
BIRADS III (%)
BIRADS IV (%)
BIRADS V (%)
32 100 16 p < 0.00001
62 100 35 p < 0.00001
20 100 10 p < 0.00001
21 100 10 p = 0.00001
14 100 8 p < 0.00001
70 100 51 p = 0.00014
100 100 100 ns
BIRADS: Breast Imaging Reporting and Data System; SM: scintimammography.
between 23 % and 33 %.18,20,21 In a recent meta-analysis by Taillefer,27 which excluded multicentre studies, the percentage of nonpalpable lesions was 41 %. However, in studies published in recent years, we noticed a tendency to increase the percentage of nonpalpable lesions studied. In their series, Tiling et al.22 found 46 % nonpalpable lesions, Khalkhali et al.,25 56 %, and Sampalis et al.,26 67 %, data which are more in accordance with the 72 % of our series. Likewise, it is interesting to note that the percentage of nonpalpable lesions present in the 2 large meta-analyses published by Taillefer rose from 30 % in 199928 to 41 % in 2005.27 Diagnosis of a palpable lesion is based on the physical examination, mammography and fine needle aspiration cytology. Management of a nonpalpable lesion is more difficult, even more so considering the low PPV of mammography,6,7 which usually leads to performing a stereotaxic biopsy, a more complex
technique with greater morbidity. This causes scars in the breast tissue, which makes it difficult to interpret subsequent mammographies. Therefore, it would be fundamentally in these nonpalpable lesions in which surgeons and/or gynecologists would need a complementary test to help better assess their nature and reduce the number of biopsies performed on benign lesions. Few studies have analyzed the type of mammographic lesions assessed by SM, although, based on the high percentages of cancer generally found in the literature, 56 % in the meta-analysis by Liberman et al.,24 we assume that they include lesions having high suspicion of malignancy. In our study, there is a predominance of lesions that are shown to be benign or probably benign (BIRADS I, II and III) in the mammography, 73.5 % of the total, while malignant or probably malignant lesions only account for 26.5 %. Similar results have been reported by Sampalis et al.
Rev Esp Med Nucl. 2007;26(3):153-9
157
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
with 72 % and 28 %, respectively.26 Thus, both studies show how SM currently tends to be used in lesions with low suspicion of malignancy. In lesions that are highly suggestive of malignancy and, to a lesser degree, in suspicious lesions, the anatomical-pathological study is generally the alternative used most in their assessment. In our experience, the percentage of cancers diagnosed in BIRADS IV and V lesions was 40 % and 100 %, respectively, while it was only 5 % in the BIRADS I, II and III. If we review the SM results, we can observe how its sensitivity, specificity, PPV, NPV and diagnostic accuracy are 83 %, 86 %, 83 %, 86 % and 85 %, respectively in the Taillefer meta-analysis.27 The main limitation of these studies, as well as in the main multicentre studies published,20,21 is that most of them study high-risk patients, which results in a high prevalence of cancer in the studies. Thus, in our results obtained in a predominately low risk population, we can observe that although the sensitivity of the technique is similar (85 %) and the PPV is even somewhat lower (80 %), there is an increase in specificity and NPV (95 % and 96 %). This finding, which reflects the high proportion of TN results, has also been observed by other authors in studies conducted in populations with a low percentage of breast cancer.25,26 We found a clear relationship between tumor size and SM sensitivity. Thus, only 50 % of T1 a-b lesions were detected by SM, which rose to 91.7 % in T1c and 100 % in T2. Similar findings have been widely documented and indicate the little usefulness of SM in lesions of less than 1 cm.18,22 If we evaluate the results according to whether the lesion is palpable or not, we see that nonpalpable lesions have lower sensitivity and PPV in the SM, although the specificity and NPV remain stable. These results are fundamentally due to the low proportion of malignant lesions present (10 %) and the smaller size of the tumors included. However, as mentioned previously, the results of SM are linked to the lesion size and not to whether it is palpable or not. The greatest advantage of SM in the clinical management of patients is in nonpalpable lesions, which have more diagnostic difficulty but a size greater than 1 cm. It should be noted that most of the nonpalpable lesions found in the mammography are larger than 1 cm. In a study conducted in our hospital on 502 nonpalpable lesions detected by mammography, 78 % were larger than 1 cm.29 158
If we analyze the results according to the BIRADS of the lesion, several points must be considered. The first is the high NPV present in all the groups, although it is greater in BIRADS I and II, and a clear increase in the PPV of SM depending on the BIRADS of the lesion. These data are a direct consequence of the proportion of cancers in each group. Likewise, the lower sensitivity of SM in BIRADS IV lesions due to the high proportion of small-sized lesions in this group should be noted. The results of SM significantly influenced the clinical management of the patients in all lesion groups except for those classified on mammography as BIRADS V. This influence is fundamentally based on the high NPV of SM, which supposes that a negative SM is highly suggestive of benignancy, as long as the lesion is larger than 1 cm. Thus, the pathological study was only performed in 16 % of lesions with negative SM, but in all the lesions with positive SM. This influence is particularly apparent in nonpalpable lesions and in those in which the behavior in the mammography is normal, benign or probably benign. This usefulness of SM has been highlighted by other authors.19,24,26,30 They have concluded that the high NPV of SM makes it very useful in the selection of lesions larger than 1 cm, in which routine biopsy of the breast is not needed but which they can follow-up with noninvasive procedures. Biopsy would be the procedure of choice in lesions under 1 cm and/or with high suspicion of malignancy on mammography, that is, BIRADS V and, in a lower proportion, BIRADS IV. It is likewise important to observe how SM is also useful in the detection of unsuspected breast cancers. SM detected 11 cancers (19 % of the total) which were not detected by mammography or that were seen as benign or probably benign lesions. As Sampalis et al.26 indicated, SM significantly increases the diagnostic accuracy of mammography, reducing the number of unnecessary biopsies and occult breast cancers. In conclusion, we can state that, in our clinical experience, SM is fundamentally performed in populations with a low prevalence of breast cancer to assess nonpalpable lesions with low suspicion of malignancy on mammography. SM has a high NPV, increases the diagnostic accuracy of mammography and has repercussion in the clinical management of patients, except in lesions classified as BIRADS V.
Rev Esp Med Nucl. 2007;26(3):153-9
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
REFERENCES 1. Freig SA. Role and evaluation of mammography and other imaging methods for breast cancer detection, diagnosis and staging. Semin Nucl Med. 1999;24:3-15. 2. Huyh PT, Jarolimek AM, Daye S. The false-negative mammogram. RadioGraphics. 1998;18:1137-54. 3. Porter PL, El-Bastawissi AY, Mandelson MT, Lin MG, Khalid N, Watney EA, et al. Breast tumor characteristics as predictors of mammographic detection: comparison of interval- and screendetected cancers. J Nucl Cancer Institute. 1999;91:2020-8. 4. Mandelson MT, Oestreicher N, Porter PL, White D, Finder CA, Taplin SH, et al. Breast density as a predictor of mammographic detection: comparison of interval- and screen-detected cancers. J Nucl Cancer Institute. 2000;92:1081-7. 5. Sickles EA. Mammographic features of 300 consecutive nonpalpable breast cancer. AJR. 1986;146:661-3. 6. Kopans D. The predictive positive value of mammography. AJR. 1992;158:521-6. 7. Kopans D, Moore RH, McCarthy KA, Hall DH, Hulka CA, Whitman GJ, et al. Positive predictive value of breast biopsy performed as a result of mammography: there is no abrupt change at age 50 years. Radiology. 1996;200:357-60. 8. Waxman A, Nagaraj N, Ashok G, Khan S, Memsic L, Yadegar J, et al. Sensitivity and specificity of 99mTc in the evaluation of primary carcinoma of the breast: comparison of palpable and nonpalpable lesions with mammography. J Nucl Med. 1994;35:22. Abstract. 9. Khalkhali I, Mena I, Jouanne E, Diggles L, Venegas R, Block J, et al. Prone scintimammography in patients with suspicion of carcinoma of the breast. J Am Coll Surg. 1994;178:491-7. 10. Khalkhali I, Mena I, Diggles L. Review of imaging techniques for the diagnosis of breast cancer: a new role of prone scintimammography using technetium-99m sestamibi. Eur J Nucl Med. 1994;21:357-62. 11. Taillefer R, Robidoux A, Lambert R, Laperrine J. Technetium-99m sestamibi prone scintimammography to detect primary breast cancer and axilary lymph node involvement. J Nucl Med. 1995;36:1758-65. 12. Khalkhali I, Villanueva-Meyer J, Edell SL, Connolly JL, Schnitt SJ, Braum JK, et al. Diagnostic accuracy of Tc-99m sestamibi breast imaging in breast cancer detection. J Nucl Med. 2000;41: 1973-9. 13. Palmedo H, Schomburg A, Grünwald F, Mallmann P, Krebs D, Biersack H. Technetium-99m-MIBI scintimammography for suspicious breast lesions. J Nucl Med. 1996;37:626-30. 14. Tabuenca O, Quirce R, Gómez-Barquín R, Blanco I, Uriarte I, Vega A, et al. La gammagrafía con MIBI-99mTc en el diagnóstico del carcinoma primario de mama en 28 pacientes con criterios mamográficos de malignidad. Rev Esp Med Nucl. 1996;15: 149-53. 15. Villanueva-Meyer J, Leonard MH Jr, Briscoe E, Cesani F, Ali SA, Rhoden S, et al. Mammoscintigraphy with technetium-99m-
16. 17.
18.
19.
20.
21.
22.
23. 24.
25.
26.
27. 28.
29.
30.
sespamibi in suspected breast cancer. J Nucl Med. 1996;37: 926-30. Waxman AD. The role of 99mTc Methoxyisobutylisonitrile in imaging breast cancer. Semin Nucl Med. 1997;27:40-54. Carril JM, Gómez-Barquín R, Quirce R, Tabuenca O, Uriarte I, Montero A. Contribution of 99mTc-MIBI scintimammography to the diagnosis of non-palpable breast lesions in relation to mammographic probability of malignancy. Anticancer Research. 1997;17:1677-82. Scopinaro F, Schillaci O, Ussif W, Nordling K, Capopferro R, De Vicentis G, et al. A three center study on the diagnostic accuracy of 99mTc-MIBI scintimammography. Anticancer Research. 1997;17:1631-4. Prats E, Aisa F, Abós MD, Villavieja L, García Y, Asenjo MJ, et al. Mammography and 99mTc-MIBI scintimammography in suspected breast cancer. J Nucl Med. 1999;40:296-301. Palmedo H, Biersack HJ, Lastoria S, Maublanc J, Prats E, Steger HE, et al. Scintimammography with technetium-99m methoxyisobutylisonitrile: results of a prospective European multicentre trial. Eur J Nucl Med. 1998;25:375-85. Prats E, Carril JM, Herranz R, Meroño E, Banzo J. Estudio multicéntrico español “Gammagrafía de mama con MIBI-99mTc”. Presentación de resultados. Rev Esp Med Nucl. 1998;17: 338-50. Tiling R, Kebler M, Untch M, Sommer H, Linke R, Hahn K. Initial evaluation of breast cancer using 99mTc-sestamibi scintimammography. Eur J Radiol. 2005;53:206-12. Liberman L, Menell JH. Breast imaging reporting and data system (BI-RADS). Radiol Clin N Am. 2002;40:409-30. Liberman M, Sampalis F, Mulder DS, Sampalis JS. Breast cancer diagnosis by scintimammography: a meta-analysis and review of the literature. Breast Cancer Research and Treatment. 2003;80:115-26. Khalkhali I, Villanueva-Meyer J, Edell SL, Connolly JL, Schnitt ST, Baum JK, et al. Diagnostic accuracy of 99mTc-sestamibi breast imaging multicenter trial results. J Nucl Med. 2000;41: 1973-9. Sampalis FS, Denis R, Picard D, Fleiszer D, Martin G, Nassif E, et al. International prospective evaluation of scintimammography with 99mTechnetium sestamibi. Am J Surg. 2003;544-9. Taillefer R. Clinical applications of 99mTc-sestamibi scintimammography. Semin Nucl Med. 2005;35:100-15. Taillefer R. The role of 99mTc-sestamibi and other conventional radiopharmaceuticals in breast cancer diagnosis. Semin Nucl Med. 1999;24:16-40. Sainz JM. Incremento del valor predictivo positivo mediante la cuantificación del riesgo radiológico en el cáncer de mama no palpable. Ed. Consejo Económico y Social de Aragón. Zaragoza. 2002. Mirzaei S, Zajicek SM, Knoll P, Lipps C, Lipp RW, Salzer H, et al. Scintimammography enhances negative predictive value of non-invasive pre-operative assessment of breast lesions. Eur J Surg Onc. 2000;26:738-41.
Rev Esp Med Nucl. 2007;26(3):153-9
159
Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient E. PRATSa, P. RAZOLAa, J.M. SAINZb, L. TARDÍNa, A. ANDRÉSa, M.D. ABÓSa, F. GARCÍAa, J. FERNANDEZb, L. VILLAVIEJAb AND J. BANZOa Departments of aNuclear Medicine and bRadiodiagnosis. Hospital Clínico Universitario Lozano Blesa. Zaragoza. Spain.
Abstract.—Objective. Establish the usefulness of scintimammography (SM) in day-to-day clinical practice. Material and methods. We have evaluated 308 SM consecutively performed in our Department. The diagnosis was established by way of biopsy or clinical and mammography follow-up (minimum 18 months). In all cases we evaluated the kind of lesions (palpable or non-palpable) and their degree of suspicion of malignancy in the mammography (BIRADS). We determined the number of fine-needle aspiration cytology (FNAC) or biopsies performed in dependence of kind of lesions, mammography and SM results. Results. Final diagnosis was cancer in 18 % of the patients, and 71 % of the lesions were non palpable. Negative predictive value (NPV) of SM was 96 %. Pathological diagnosis was performed in 100 % of lesions with SM+ and 16 % of lesions with SM–. According to the kind of lesion, FNAC and/or biopsy was performed in 62 % of palpable lesions and in 20 % of non-palpable lesions; and according to the mammography in 21 % of lesions BIRADS I-II, 14 % in BIRADS III, 70 % in BIRADS IV and 100 % in BIRADS V. The number of FNAC and/or biopsies performed is significantly higher in all cases when SM is positive (p > 0.001), excepting in BIRADS V lesions. Conclusions. In day-to-day clinical practice SM is generally performed in low prevalence of breast cancer population, in the evaluation of non-palpable lesions with a low suspicion of malignancy. SM has a high NPV, increase the diagnostic accuracy of the mammography and has repercussion on patient management, except in BIRADS V lesions.
Material y método. Hemos revisado 308 GM realizadas en nuestro servicio. El diagnóstico se estableció mediante citología y/o biopsia o por seguimiento clínico/mamográfico superior a 18 meses. En todos los casos se valoró el tipo de lesión (palpable o no) y su sospecha de malignidad en mamografía (BIRADS). Se determinó el número de punciones aspiraciones con aguja fina (PAAF) y/o biopsias realizadas en dependencia del resultado de la GM, tipo de lesión y mamografía. Resultados. El diagnóstico final fue de cáncer en el 18 % de las pacientes, siendo no palpables el 71 % de las lesiones. El valor predictivo negativo (VPN) de la GM fue del 96 %. Se realizó diagnóstico anatomopatológico en el 100 % de lesiones con GM positiva y en el 16 % de lesiones con GM negativa. Dependiendo del tipo de lesión, se practicó PAAF y/o biopsia en el 62 % de lesiones palpables y en el 20 % de las no palpables; según la mamografía, en el 21 % de lesiones BIRADS I-II, 14 % BIRADS III, 70 % BIRADS IV y 100 % BIRADS V. El número de estudios anatomopatológicos practicados fue significativamente mayor en todos los grupos con GM positiva (p > 0,001), excepto en BIRADS V. Conclusiones. En la práctica clínica, la GM se realiza fundamentalmente en poblaciones con baja prevalencia de cáncer, en lesiones no palpables de baja sospecha de malignidad. La GM tiene un alto VPN, incrementa la seguridad diagnóstica de la mamografía y repercute en el manejo de las pacientes, excepto en lesiones BIRADS V. PALABRAS CLAVE: gammagrafía de mama, 99mTc-MIBI, cáncer de mama.
KEY WORDS: scintimammography, 99mTc-MIBI, breast cancer. GAMMAGRAFÍA DE MAMA CON 99MTC-MIBI EN LA PRÁCTICA CLÍNICA. REPERCUSIÓN EN EL MANEJO CLÍNICO DEL PACIENTE Resumen.—Objetivo. Valorar el papel de la gammagrafía de mama (GM) en la práctica clínica. Correspondence: E. PRATS. Servicio de Medicina Nuclear. Hospital Clínico Universitario Lozano Blesa. Avda. San Juan Bosco, 15. 50009 Zaragoza. Spain. E-mail: [email protected] Received: 18-10-06 Accepted: 24-01-07
Mammography is the most useful procedure in the diagnosis of breast cancer, both in the screening examination and when faced with clinical suspicion of the disease.1 However, it has some limitations. Its sensitivity may be reduced in certain situations, such as dense breasts or presence of scar tissue, the cancer rate of the interval described varying between 3 % and 20 %.1-4 Furthermore, mammography is a relatively unspecific diagnostic technique which often does not distinguish malignant lesions from benign pathology.5 Its positive predictive value (PPV) varies between 20 % and 40 %, which leads to an elevated number of biopsies being performed on benign lesions.6,7
Rev Esp Med Nucl. 2007;26(3):153-9
153
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
99m
Tc-MIBI scintimammography (SM) was introduced into clinical practice in 1994 after studies by Waxman et al.8 and Khalkhali et al.9,10 Since then, many papers analyzing the methodology, results or indications of this technique have been published.11-22 However, there is still little information on the real role of this technique in daily clinical practice, evaluating the type of lesions studied, their results or their repercussion in the clinical management of the patient. SM has been used in our hospital as a diagnostic technique complementary to mammography since the end of the last decade. In this setting, the objectives that we proposed in this paper, on assessment of a large series of patients studied in the same centre following a routine diagnostic methodology, were the following: 1) To analyze the study patient sample, behavior of the lesions in the physical examination and mammography and the SM results, based on global results and depending on the different types of lesions, and 2) To assess the repercussion of SM on the clinical management of patients.
MATERIAL AND METHODS
Patients and study design
We retrospectively analyzed 350 SMs performed consecutively in our Nuclear Medicine Department. The patients were referred from different departments of the hospital or from attached specialty centers. Study inclusion criteria were: 1) Availability of the patient’s physical examination data, assessing whether the lesion was palpable or not, and a recent mammography and SM; and 2) Establishment of a definitive diagnosis by cytology and/or biopsy. When no anatomical-pathological study was available, the diagnosis of “benignancy” was made by clinical/mammographic follow-up for a period of at least 18 months. In accordance with these criteria, the total number of SMs included was 308. All patients were women with a mean age of 54 years (range 26-84 years). Mammography
Mammography was performed with cranial-caudal and mediolateral oblique views in all cases, taking images in the lateral view, magnified or compressed when the radiologist considered it necessary. In all the mammographies, the presentation patterns and 154
diagnostic category were assessed in accordance with the American College of Radiology Breast Imaging Reporting and Data System (BIRADS). The diagnostic categories of BIRADS are: I (normal), II (benign), III (probably benign), IV (suspicion) and V (high suspicion of malignancy).23 Scintimammography
The SM was carried out 5 minutes after the intravenous injection of 740 MBq of 99mTc-MIBI (hexakis 2 methoxyisobutyl isonitrile, Cardiolite, Bristol Mayers) in a dorsal vein of the foot. The radiotracer was administered through a butterfly needle and washed with 5-10 cc of physiological serum. SMs were acquired using Elscint SP6, Elscint Helix SPX and Millenium VG General Electric gammacameras. Lateral images of both breasts in anterior and prone decubitus positions were obtained using a specially designed table with lateral openings which allow the breast to fall freely, thus obtaining the maximum separation between deep breast tissue and organs with physiological uptake of 99mTc-MIBI, such as the liver or myocardium. The acquisition time per image was 10 minutes using a high-resolution collimator and 256 × 256 matrix. All focal or multifocal uptake of the radiotracer that was greater than the breast background was considered positive, regardless of the degree of uptake. Diffuse uptakes were not considered suggestive of malignancy. All SMs were reviewed independently by two Nuclear Medicine specialists. When they disagreed, the opinion of a third specialist was sought. Statistical analysis
We used the χ2 test with Yate’s correction to study if there were statistically significant differences between the number of biopsies and/or FNAC performed according to the result of the SM, both in the global sample and depending on the type of lesion studied (palpable or not) and its diagnostic category on mammography (BIRADS). We considered values of p < 0.01 as significant.
RESULTS
Diagnoses were established with anatomicalpathological criteria (cytology and/or biopsy) in
Rev Esp Med Nucl. 2007;26(3):153-9
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
99 patients (32 %). Presence of breast cancer was demonstrated in 56 patients (18 % of the total): 44 invasive ductal carcinomas (IDC), 9 invasive lobular carcinomas (ILC) and 3 in situ ductal carcinomas (DC in situ); no malignant pathology was found in the remaining 43 (14 %) patients: 24 fibrocystic breast disease (FBD), 7 fibroadenomas, 5 post-surgical scars, 2 fat necrosis, 1 scar associated with an inflammatory process, 1 hamartoma, 1 post-traumatic necrosis and 2 with no detectable breast pathology. Malignancy was excluded in the remaining 209 patients (68 %) as no breast pathology was demonstrated during the clinical/ mammographic follow-up of at least 18 months. The number of benign lesions to be studied was therefore 252 (82 %). Eighty-nine lesions
were detected with palpation (29 %) while 219 lesions were not palpable (71 %). The mammographies were assessed in accordance with the following BIRADS categories: I in 18 cases (6 %), II in 28 (9 %), III in 180 (58.5 %), IV in 63 (20.5 %) and V in 19 (6 %). SM was positive in 60 lesions (48 true positive [TP], 12 false positive [FP]) and negative in 248 (240 true negative [TN], 8 false negative [FN]) (Figs. 1-3). Thus, the global results obtained were: sensitivity 85.7 %, specificity 95.2 %, PPV 80 %, negative predictive value (NPV) 96.8 % and diagnostic accuracy 93.5 %. Six of the FN results occurred in lesions of less than 1 cm in diameter (4 IDC, 1 ILC and 1 DC in situ), while the remaining 2 neoplasia
A
B
FIG. 1.—Patient with palpable nodule in inferior-external right breast quadrant. Mammography (A): Extremely dense breast without obvious nodules. Scintimammography (B): Uptake focus coinciding with palpable lesion. Definitive diagnosis: Invasive ductal carcinoma (biopsy).
A
B
FIG. 2.—Fifty-five year-old woman without palpable breast lesion. Mammography (A): Asymmetric density increase in left breast retroareolar region (BIRADS III) with bilateral nodules of benign aspect. Scintimammography (B): Normal. Clinical/radiological follow-up: > 18 months without showing breast pathology. Rev Esp Med Nucl. 2007;26(3):153-9
155
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
A
B
FIG. 3.—Patient operated on for invasive ductal carcinoma 2 years ago. Mammography (A): Post-surgical scar in upper right breast quadrant (BIRADS III). Scintimammography (B): Normal. Clinical/radiological follow-up > 18 months, without showing breast pathology.
not detected in the SM were 2 IDC of between 1 and 2 cm. The anatomical-pathological results corresponding to the FP results were: 7 FBD, 1 fibroadenoma, 1 hamartoma, 1 fat necrosis, 1 post-traumatic necrosis and 1 post-surgical scar with associated inflammatory process. In the 89 palpable lesions (Fig. 1), SM was positive in 42.5 % of the patients (33 TP, 5 FP), and negative in 57.5 % (50 TN, 1 FN), while in the 219 nonpalpable lesions (Figs. 2 and 3), SM was positive in 10 % (15 TP, 7 FP) and negative in 90 % (190 TN, 7 FN). Global results and the results of palpable and nonpalpable lesions are shown in Table 1. If we analyze the size of the breast cancers included in our study, we find 12 T1 a-b (< 1 cm), 24 T1c (1-2 cm) and 20 T2 (> 2 cm). SM detected 6 T1a-b (sensitivity 50 %), 22 T1c (sensitivity 91.7 %) and 20 T2 (sensitivity 100 %). The number of breast cancers found in each BIRADS class was: 2 in BIRADS I (9 %), 1 in BIRADS II (4 %), 9 in BIRADS III (5 %), 25 in BIRADS IV (40 %) and 19 in BIRADS V (100 %). Correlation of SM results with the BIRADS classes shows that the SM was positive in 13 % (3 TP, 3 FP) and negative in 87 % (40 TN) of the 46 BIRADS I or II lesions. Among the 180 BIRADS III lesions (Figs. 2 and 3), the SM was positive in 6.6 % (8 TP, 4 FP) and negative in 93.4 % (167 TN, 1 FN). Among the 63 BIRADS IV, the SM was positive in 38 % 156
(19 TP, 5 FP) and negative in 62 % (33 TN, 6 FN). Finally, among the 19 BIRADS V lesions, the SM was positive in 94.7 % (18 TP) and negative 3.3 % (1 FN). The global results of the SM and the relationship with each type of mammographic lesions according to the BIRADS criteria are shown in Table 2. The SM allowed detection of 11 breast cancers in which the mammography was not suspicious of malignancy (3 BIRADS I/II and 8 BIRADS III). The pathological studies conducted according to SM results for the global sample and according to lesion type (palpable or nonpalpable) and its diagnostic classification on mammography are described in Table 3. As can be seen in the Table, the number of biopsies and/or cytology performed is significantly greater in all the groups when the SM is positive (p < 0.00001 in the global sample, palpable lesions, nonpalpable lesions and lesions with BIRADS I, II and II mammography, and p = 0.00014 in BIRADS IV), except in BIRADS V.
DISCUSSION
Some data of interest regarding the type of lesions studied, their presentation on mammography, clinical results of the SM in different types of lesions and how the SM influences the clinical management of patients can be obtained by analyzing our results. In first place, we can note the high proportion of benign lesions studied with SM. A cancer diagnosis was established in only 56 (18 %) of all the lesions while 252 (82 %) were benign. The percentage of malignant lesions was less than that generally found in the literature. In a meta-analysis and review of the SM results that included 64 articles published up to 1999,24 56 % of the lesions were diagnosed as cancer. Equally, if we analyze multicentre studies, we see that the proportion of malignant lesions varied between 40 % and 84 %.18,25 In a recent study published by Sampalis et al26 which included 1,243 patients, only 16 % had malignant pathology, which is more in accordance with our study. If we consider that SM is a technique complementary to mammography, its use outside of research studies or to assess the response to chemotherapy would be little justified in population groups with high percentages of breast cancer. In most publications on SM, there is a clear predominance of examinations performed in palpable lesions. In the first studies, nonpalpable lesions varied
Rev Esp Med Nucl. 2007;26(3):153-9
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
Table 1 GLOBAL RESULTS OF THE SCINTIMAMMOGRAPHY IN PALPABLE AND NONPALPABLE LESIONS Lesion type
Sensitivity (%)
Specificity(%)
Accuracy(%)
PPV(%)
NPV(%)
Palpable (n = 89) Nonpalpable (n = 219) Total (n = 308)
97 68 85
94 96 95
93 93 93
91 68 80
98 96 96
NPV: negative predictive value; PPV: positive predictive value.
Table 2 RESULTS OF THE SCINTIMAMMOGRAPHY ACCORDING TO THE BIRADS OF THE LESION BIRADS
Sensitivity (%)
Specificity (%)
Accuracy (%)
PPV (%)
NPV (%)
I-II (n = 46) III (n = 180) IV (n = 63) V (n = 19) Total
100 89 76 94 85
93 97 87 – 95
93 97 82 94 93
50 66 79 100 80
100 99 85 –0 96
BIRADS: Breast Imaging Reporting and Data System; NPV: negative predictive value; PPV: positive predictive value.
Table 3 PATHOLOGICAL STUDIES PERFORMED ACCORDING TO SCINTIMAMMOGRAPHY RESULTS
Global SM+ SM– p
Global (%)
Palpable lesions (%)
Nonpalpable lesions (%)
BIRADS I-II (%)
BIRADS III (%)
BIRADS IV (%)
BIRADS V (%)
32 100 16 p < 0.00001
62 100 35 p < 0.00001
20 100 10 p < 0.00001
21 100 10 p = 0.00001
14 100 8 p < 0.00001
70 100 51 p = 0.00014
100 100 100 ns
BIRADS: Breast Imaging Reporting and Data System; SM: scintimammography.
between 23 % and 33 %.18,20,21 In a recent meta-analysis by Taillefer,27 which excluded multicentre studies, the percentage of nonpalpable lesions was 41 %. However, in studies published in recent years, we noticed a tendency to increase the percentage of nonpalpable lesions studied. In their series, Tiling et al.22 found 46 % nonpalpable lesions, Khalkhali et al.,25 56 %, and Sampalis et al.,26 67 %, data which are more in accordance with the 72 % of our series. Likewise, it is interesting to note that the percentage of nonpalpable lesions present in the 2 large meta-analyses published by Taillefer rose from 30 % in 199928 to 41 % in 2005.27 Diagnosis of a palpable lesion is based on the physical examination, mammography and fine needle aspiration cytology. Management of a nonpalpable lesion is more difficult, even more so considering the low PPV of mammography,6,7 which usually leads to performing a stereotaxic biopsy, a more complex
technique with greater morbidity. This causes scars in the breast tissue, which makes it difficult to interpret subsequent mammographies. Therefore, it would be fundamentally in these nonpalpable lesions in which surgeons and/or gynecologists would need a complementary test to help better assess their nature and reduce the number of biopsies performed on benign lesions. Few studies have analyzed the type of mammographic lesions assessed by SM, although, based on the high percentages of cancer generally found in the literature, 56 % in the meta-analysis by Liberman et al.,24 we assume that they include lesions having high suspicion of malignancy. In our study, there is a predominance of lesions that are shown to be benign or probably benign (BIRADS I, II and III) in the mammography, 73.5 % of the total, while malignant or probably malignant lesions only account for 26.5 %. Similar results have been reported by Sampalis et al.
Rev Esp Med Nucl. 2007;26(3):153-9
157
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
with 72 % and 28 %, respectively.26 Thus, both studies show how SM currently tends to be used in lesions with low suspicion of malignancy. In lesions that are highly suggestive of malignancy and, to a lesser degree, in suspicious lesions, the anatomical-pathological study is generally the alternative used most in their assessment. In our experience, the percentage of cancers diagnosed in BIRADS IV and V lesions was 40 % and 100 %, respectively, while it was only 5 % in the BIRADS I, II and III. If we review the SM results, we can observe how its sensitivity, specificity, PPV, NPV and diagnostic accuracy are 83 %, 86 %, 83 %, 86 % and 85 %, respectively in the Taillefer meta-analysis.27 The main limitation of these studies, as well as in the main multicentre studies published,20,21 is that most of them study high-risk patients, which results in a high prevalence of cancer in the studies. Thus, in our results obtained in a predominately low risk population, we can observe that although the sensitivity of the technique is similar (85 %) and the PPV is even somewhat lower (80 %), there is an increase in specificity and NPV (95 % and 96 %). This finding, which reflects the high proportion of TN results, has also been observed by other authors in studies conducted in populations with a low percentage of breast cancer.25,26 We found a clear relationship between tumor size and SM sensitivity. Thus, only 50 % of T1 a-b lesions were detected by SM, which rose to 91.7 % in T1c and 100 % in T2. Similar findings have been widely documented and indicate the little usefulness of SM in lesions of less than 1 cm.18,22 If we evaluate the results according to whether the lesion is palpable or not, we see that nonpalpable lesions have lower sensitivity and PPV in the SM, although the specificity and NPV remain stable. These results are fundamentally due to the low proportion of malignant lesions present (10 %) and the smaller size of the tumors included. However, as mentioned previously, the results of SM are linked to the lesion size and not to whether it is palpable or not. The greatest advantage of SM in the clinical management of patients is in nonpalpable lesions, which have more diagnostic difficulty but a size greater than 1 cm. It should be noted that most of the nonpalpable lesions found in the mammography are larger than 1 cm. In a study conducted in our hospital on 502 nonpalpable lesions detected by mammography, 78 % were larger than 1 cm.29 158
If we analyze the results according to the BIRADS of the lesion, several points must be considered. The first is the high NPV present in all the groups, although it is greater in BIRADS I and II, and a clear increase in the PPV of SM depending on the BIRADS of the lesion. These data are a direct consequence of the proportion of cancers in each group. Likewise, the lower sensitivity of SM in BIRADS IV lesions due to the high proportion of small-sized lesions in this group should be noted. The results of SM significantly influenced the clinical management of the patients in all lesion groups except for those classified on mammography as BIRADS V. This influence is fundamentally based on the high NPV of SM, which supposes that a negative SM is highly suggestive of benignancy, as long as the lesion is larger than 1 cm. Thus, the pathological study was only performed in 16 % of lesions with negative SM, but in all the lesions with positive SM. This influence is particularly apparent in nonpalpable lesions and in those in which the behavior in the mammography is normal, benign or probably benign. This usefulness of SM has been highlighted by other authors.19,24,26,30 They have concluded that the high NPV of SM makes it very useful in the selection of lesions larger than 1 cm, in which routine biopsy of the breast is not needed but which they can follow-up with noninvasive procedures. Biopsy would be the procedure of choice in lesions under 1 cm and/or with high suspicion of malignancy on mammography, that is, BIRADS V and, in a lower proportion, BIRADS IV. It is likewise important to observe how SM is also useful in the detection of unsuspected breast cancers. SM detected 11 cancers (19 % of the total) which were not detected by mammography or that were seen as benign or probably benign lesions. As Sampalis et al.26 indicated, SM significantly increases the diagnostic accuracy of mammography, reducing the number of unnecessary biopsies and occult breast cancers. In conclusion, we can state that, in our clinical experience, SM is fundamentally performed in populations with a low prevalence of breast cancer to assess nonpalpable lesions with low suspicion of malignancy on mammography. SM has a high NPV, increases the diagnostic accuracy of mammography and has repercussion in the clinical management of patients, except in lesions classified as BIRADS V.
Rev Esp Med Nucl. 2007;26(3):153-9
Prats E et al. Scintimammography with 99mTc-MIBI in clinical practice. Repercussion on the clinical management of the patient
REFERENCES 1. Freig SA. Role and evaluation of mammography and other imaging methods for breast cancer detection, diagnosis and staging. Semin Nucl Med. 1999;24:3-15. 2. Huyh PT, Jarolimek AM, Daye S. The false-negative mammogram. RadioGraphics. 1998;18:1137-54. 3. Porter PL, El-Bastawissi AY, Mandelson MT, Lin MG, Khalid N, Watney EA, et al. Breast tumor characteristics as predictors of mammographic detection: comparison of interval- and screendetected cancers. J Nucl Cancer Institute. 1999;91:2020-8. 4. Mandelson MT, Oestreicher N, Porter PL, White D, Finder CA, Taplin SH, et al. Breast density as a predictor of mammographic detection: comparison of interval- and screen-detected cancers. J Nucl Cancer Institute. 2000;92:1081-7. 5. Sickles EA. Mammographic features of 300 consecutive nonpalpable breast cancer. AJR. 1986;146:661-3. 6. Kopans D. The predictive positive value of mammography. AJR. 1992;158:521-6. 7. Kopans D, Moore RH, McCarthy KA, Hall DH, Hulka CA, Whitman GJ, et al. Positive predictive value of breast biopsy performed as a result of mammography: there is no abrupt change at age 50 years. Radiology. 1996;200:357-60. 8. Waxman A, Nagaraj N, Ashok G, Khan S, Memsic L, Yadegar J, et al. Sensitivity and specificity of 99mTc in the evaluation of primary carcinoma of the breast: comparison of palpable and nonpalpable lesions with mammography. J Nucl Med. 1994;35:22. Abstract. 9. Khalkhali I, Mena I, Jouanne E, Diggles L, Venegas R, Block J, et al. Prone scintimammography in patients with suspicion of carcinoma of the breast. J Am Coll Surg. 1994;178:491-7. 10. Khalkhali I, Mena I, Diggles L. Review of imaging techniques for the diagnosis of breast cancer: a new role of prone scintimammography using technetium-99m sestamibi. Eur J Nucl Med. 1994;21:357-62. 11. Taillefer R, Robidoux A, Lambert R, Laperrine J. Technetium-99m sestamibi prone scintimammography to detect primary breast cancer and axilary lymph node involvement. J Nucl Med. 1995;36:1758-65. 12. Khalkhali I, Villanueva-Meyer J, Edell SL, Connolly JL, Schnitt SJ, Braum JK, et al. Diagnostic accuracy of Tc-99m sestamibi breast imaging in breast cancer detection. J Nucl Med. 2000;41: 1973-9. 13. Palmedo H, Schomburg A, Grünwald F, Mallmann P, Krebs D, Biersack H. Technetium-99m-MIBI scintimammography for suspicious breast lesions. J Nucl Med. 1996;37:626-30. 14. Tabuenca O, Quirce R, Gómez-Barquín R, Blanco I, Uriarte I, Vega A, et al. La gammagrafía con MIBI-99mTc en el diagnóstico del carcinoma primario de mama en 28 pacientes con criterios mamográficos de malignidad. Rev Esp Med Nucl. 1996;15: 149-53. 15. Villanueva-Meyer J, Leonard MH Jr, Briscoe E, Cesani F, Ali SA, Rhoden S, et al. Mammoscintigraphy with technetium-99m-
16. 17.
18.
19.
20.
21.
22.
23. 24.
25.
26.
27. 28.
29.
30.
sespamibi in suspected breast cancer. J Nucl Med. 1996;37: 926-30. Waxman AD. The role of 99mTc Methoxyisobutylisonitrile in imaging breast cancer. Semin Nucl Med. 1997;27:40-54. Carril JM, Gómez-Barquín R, Quirce R, Tabuenca O, Uriarte I, Montero A. Contribution of 99mTc-MIBI scintimammography to the diagnosis of non-palpable breast lesions in relation to mammographic probability of malignancy. Anticancer Research. 1997;17:1677-82. Scopinaro F, Schillaci O, Ussif W, Nordling K, Capopferro R, De Vicentis G, et al. A three center study on the diagnostic accuracy of 99mTc-MIBI scintimammography. Anticancer Research. 1997;17:1631-4. Prats E, Aisa F, Abós MD, Villavieja L, García Y, Asenjo MJ, et al. Mammography and 99mTc-MIBI scintimammography in suspected breast cancer. J Nucl Med. 1999;40:296-301. Palmedo H, Biersack HJ, Lastoria S, Maublanc J, Prats E, Steger HE, et al. Scintimammography with technetium-99m methoxyisobutylisonitrile: results of a prospective European multicentre trial. Eur J Nucl Med. 1998;25:375-85. Prats E, Carril JM, Herranz R, Meroño E, Banzo J. Estudio multicéntrico español “Gammagrafía de mama con MIBI-99mTc”. Presentación de resultados. Rev Esp Med Nucl. 1998;17: 338-50. Tiling R, Kebler M, Untch M, Sommer H, Linke R, Hahn K. Initial evaluation of breast cancer using 99mTc-sestamibi scintimammography. Eur J Radiol. 2005;53:206-12. Liberman L, Menell JH. Breast imaging reporting and data system (BI-RADS). Radiol Clin N Am. 2002;40:409-30. Liberman M, Sampalis F, Mulder DS, Sampalis JS. Breast cancer diagnosis by scintimammography: a meta-analysis and review of the literature. Breast Cancer Research and Treatment. 2003;80:115-26. Khalkhali I, Villanueva-Meyer J, Edell SL, Connolly JL, Schnitt ST, Baum JK, et al. Diagnostic accuracy of 99mTc-sestamibi breast imaging multicenter trial results. J Nucl Med. 2000;41: 1973-9. Sampalis FS, Denis R, Picard D, Fleiszer D, Martin G, Nassif E, et al. International prospective evaluation of scintimammography with 99mTechnetium sestamibi. Am J Surg. 2003;544-9. Taillefer R. Clinical applications of 99mTc-sestamibi scintimammography. Semin Nucl Med. 2005;35:100-15. Taillefer R. The role of 99mTc-sestamibi and other conventional radiopharmaceuticals in breast cancer diagnosis. Semin Nucl Med. 1999;24:16-40. Sainz JM. Incremento del valor predictivo positivo mediante la cuantificación del riesgo radiológico en el cáncer de mama no palpable. Ed. Consejo Económico y Social de Aragón. Zaragoza. 2002. Mirzaei S, Zajicek SM, Knoll P, Lipps C, Lipp RW, Salzer H, et al. Scintimammography enhances negative predictive value of non-invasive pre-operative assessment of breast lesions. Eur J Surg Onc. 2000;26:738-41.
Rev Esp Med Nucl. 2007;26(3):153-9
159