3304jum557-712online_Layout 1 3/19/14 11:30 AM Page 585
ORIGINAL RESEARCH
Virtual Touch Tissue Imaging on Acoustic Radiation Force Impulse Elastography A New Technique for Differential Diagnosis Between Benign and Malignant Thyroid Nodules Yi-Feng Zhang, MD, Yong He, MD, Hui-Xiong Xu, MD, PhD, Xiao-Hong Xu, MD, Chang Liu, MD, Le-Hang Guo, MD, Lin-Na Liu, MD, Jun-Mei Xu, MD Objectives—Acoustic radiation force impulse elastography is a newly developed ultrasound elasticity imaging technique that included both Virtual Touch tissue quantification and Virtual Touch tissue imaging (VTI; Siemens Medical Solutions, Mountain View, CA). This study aimed to evaluate the usefulness of VTI in differentiating malignant from benign thyroid nodules. Methods—This study included 192 consecutive patients with thyroid nodules (n = 219) who underwent surgery for compressive symptoms or suspicion of malignancy. Tissue stiffness on VTI elastography was scored from 1 (soft) to 6 (hard). The VTI scores between malignant and benign thyroid nodules were compared. The intraobserver and interobserver agreement for VTI elastography was also assessed. Received March 15, 2013, from the Department of Medical Ultrasound, Shanghai Tenth People’s Hospital, Tenth People’s Hospital of Tongji University, Shanghai, China (Y.-F.Z., Y.H., H.X.X., C.L., L.-H.G., L.-N.L., J.-M.X.); Department of Ultrasound, First College of Clinical Medical Sciences, China Three Gorges University, Yichang, China (Y.H.); and Department of Ultrasound, Affiliated Hospital of Guangdong Medical College, Zhanjiang, China (X.-H.X.). Revision requested April 25, 2013. Revised manuscript accepted for publication July 29, 2013. This work was supported in part by grant NCET-06-0723 from the Chinese Ministry of Education and the Shanghai Talent Development Project from the Shanghai Human Resources and Social Security Bureau. Drs Zhang and He contributed equally to the manuscript. Address correspondence to Hui-Xiong Xu, MD, PhD, Department of Medical Ultrasound, Shanghai Tenth People’s Hospital , Tenth People's Hospital of Tongji University, 301 Yanchangzhong Rd, 200072 Shanghai, China. E-mail: [email protected] Abbreviations
ARFI, acoustic radiation force impulse; Az , area under the curve; FOV, field of view; NPV, negative predictive value; PPV, positive predictive value; VTI, Virtual Touch tissue imaging; VTQ, Virtual Touch tissue quantification doi:10.7863/ultra.33.4.585
Results—On VTI elastography: score 1 was found in 84 nodules (all benign); score 2 in 37 nodules (3 papillary carcinomas and 34 benign nodules); score 3 in 25 nodules (1 medullary carcinoma, 6 papillary carcinomas, and 18 benign nodules); score 4 in 53 nodules (50 papillary carcinomas and 3 benign nodules); score 5 in 17 nodules (14 papillary carcinomas and 3 benign nodules); and score 6 in 3 nodules (all papillary carcinomas). A VTI elasticity score of 4 or greater was highly predictive of malignancy (P < .01), and the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 87.0% (67 of 77), 95.8% (136 of 142), 91.8% (67 of 73), 93.1% (136 of 146), and 92.7% (203 of 219), respectively. The κ values were 0.69 for intraobserver agreement and 0.85 for interobserver agreement. Conclusions—Virtual Touch tissue elasticity imaging has great potential as an adjunctive tool combined with conventional sonography for differential diagnosis between benign and malignant thyroid nodules. Key Words—acoustic radiation force impulse; superficial structures; thyroid nodule; ultrasound elastography; Virtual Touch tissue imaging
T
hyroid cancer is the most common type of endocrinerelated cancer, and its incidence has risen 2.4-fold over the last 30 years.1–3 The discrimination between malignant and benign thyroid nodules is of paramount important, since the management and prognosis for them varies greatly.4–6 Palpation is the oldest and most frequently used screening method for detecting thyroid gland tumors. In general, almost 5% of the adult population has a palpable thyroid gland nodule.7 The palpation becomes very difficult for clinicians when the nodule is small or located deeply in the thyroid tissue. As a noninvasive, inexpensive, accurate, and repeatable modality, high-frequency (7.5–13 MHz) sonography has been regarded as the first-line imaging modality for thyroid disease. With conven-
©2014 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2014; 33:585–595 | 0278-4297 | www.aium.org
3304jum557-712online_Layout 1 3/19/14 11:30 AM Page 586
Zhang et al—Virtual Touch Tissue Imaging of Thyroid Nodules
tional sonography, the following sonographic features are risk factors for malignancy of a thyroid nodule: solitary nodule, hypoechoic nodule, blurred margin, “taller-thanwide” shape, and microcalcifications.8,9 However, some authors have argued that the echogenicity, shape, border, and internal vascularity did not show any significant differences between benign and malignant nodules.10 To further enhance the diagnostic ability of sonography, ultrasound elastography, including static and real-time elastography, has been introduced into the clinic with an aim of improving the diagnosis of thyroid cancer. Real-time elastography is one of the most widely used elastographic techniques, in which calculation of the tissue elasticity distribution is performed in real time, and the examination results are represented as color-coded images over the conventional B-mode images. The increase in tissue stiffness has been proven to be associated with a higher risk of malignancy.11,12 As for real-time elastography, the sensitivity and specificity for the diagnosis of malignant thyroid nodules were reported to be 86% to 92% and 89% to 97%, respectively.13–17 However, because of the characteristics of external mechanical pressure, individual differences may be present.18 Acoustic radiation force impulse (ARFI) is a newly introduced ultrasound-based elasticity imaging technique that has been applied to the thyroid in recent years.19–24 In ARFI imaging, tissue is mechanically excited by using short-duration acoustic pulses with a fixed transmit frequency to generate localized tissue displacement. In general, ARFI is between 80 and 200 microseconds. The mechanical index value of the ARFI region of interest in the Virtual Touch tissue imaging (VTI; Siemens Medical Solutions, Mountain View, CA) mode reads in the range of 1.0 to 1.6 when the region of interest is placed at a depth range of 0.5 to 8 cm. The displacement results in shear wave propagation away from the region of excitation, and the shear wave speed of the tissue is proportional to the square root of tissue elasticity. By calculating the shear wave velocity, quantitative evaluation of the tissue stiffness is available, which is named virtual touch tissue quantification (VTQ; Siemens Medical Solutions). On the other hand, the strain change under the push pulse is named VTI, which uses an algorithm that is similar to that of conventional ultrasound elastography and is displayed as a gray scale image. A VTI image is a grayscale display of relative tissue stiffness in a user-defined region of interest. This information is computed by examining the displacements of tissue elements in response to an acoustic push pulse. When a stress (longitudinal force) is applied to the tissues in the region of interest, they may behave differently, with some having greater deformation than others. By compar-
586
ing the baseline and deformed information, individual tissue elements may be labeled by their relative stiffness. A bright shade indicates relatively soft (elastic) tissue, whereas a darker shade indicates relatively stiff (nonelastic) tissue. Therefore, VTI is a qualitative mapping of ARFI, and VTQ is a quantitative estimate of tissue stiffness. Until now, VTQ has been used for the diagnosis of thyroid nodules, and the range of specificity for the diagnosis of malignant thyroid nodules was 82.2% to 95%,19–24 whereas the diagnostic value of VTI has not been well established. In this study, 219 thyroid nodules in 192 patients were assessed with VTI elastography with an aim of evaluating the usefulness of VTI in the diagnosis of thyroid nodules.
Materials and Methods Patients From April 2011 to September 2012, 192 consecutive patients with thyroid nodules were included in this study. There were 141 female and 51 male patients with a mean age ± SD of 49 ± 12 years (range, 16–72 years). Seventyeight patients had a single nodule, and 114 had multiple nodules. All of the patients underwent sonographic and VTI examinations, and the nature of all of the target nodules was pathologically confirmed after surgery. The inclusion criteria of the target nodules were as follows: (1) nodule diameter greater than 0.5 cm; (2) solid or almost solid (≤25% cystic) nodules; (3) no treatment performed on the nodules; (4) enough thyroid tissue surrounding the nodule at the same depth; and (5) thyroid surgery performed after sonography. Nodules showing obvious sonographic features of benignity, such as cystic lesions and spongiform echogenicity, were excluded from analysis. For the patients with multiple nodules, the nodules that were suspicious of malignancy (based on conventional sonographic findings) or the largest solid ones were selected for analysis. Finally, 1 nodule in each patient from the 78 patients with a single nodule (including 54 benign and 24 malignant nodules), 1 nodule in each from 89 patients with multiple nodules (including 58 benign and 31 malignant nodules), 2 nodules in each from 23 patients with multiple nodules (including 12 with 2 benign nodules in each, 10 with 2 malignant nodules in each, and 1 with 1 benign and 1 malignant nodule), and 3 nodules in each from 2 patients with multiple nodules (including 1 with 3 benign nodules, and 1 with 1 malignant and 2 benign nodules) were included in this study (Figure 1); thus, in total, 219 thyroid nodules were observed, and the sonographic characteristics are presented in Table 1. The diameter of the nodules ranged from 0.5 to 6.0 cm (mean, 1.9 ± 1.1 cm).
J Ultrasound Med 2014; 33:585–595
3304jum557-712online_Layout 1 3/19/14 11:30 AM Page 587
Zhang et al—Virtual Touch Tissue Imaging of Thyroid Nodules
Figure 1. Patients with thyroid nodules included in the study; b indicates benign; m, malignant, n, number of nodules; and pt, patient.
All of the nodules in this study underwent surgery. The indications for thyroidectomy were as follows: (1) confirmed malignancy by fine-needle aspiration (n = 4); (2) high suspicion of malignancy by sonography or fineneedle aspiration (n = 93); and (3) compressive symptoms caused by the large nodules or associated large nodules (n = 122). In the patients with compressive symptoms, 74 large nodules causing compressive symptoms were included for VTI analysis and were finally removed. For the remaining 48 small nodules, they were removed together with the associated large nodules when the nodules were present in the same lobe. These 48 nodules were included for VTI analysis because they were solid or hypoechoic on sonography, whereas the associated large nodules were not included for VTI analysis because of obvious sonographic features for benignity, such as cysts and a spongiform echo texture (Figure 2). The criteria for fine-needle aspiration were as follows: (1) patients with a high-risk history and nodule size larger than 5 mm; (2) microcalcifications present in nodules and nodule size of 10 mm or larger; (3) solid nodule and hypoechoic nodules and nodule size larger than 10 mm; (4) solid and isoechoic or hyperechoic nodules and nodule size of 10 to 15 mm or larger; and (5) mixed cystic-solid nodules with any suspicious sonographic features and nodule size of 15–20 mm or larger. Fine-needle aspiration was performed with a 22-gauge needle attached to a 5-mL disposable syringe by a freehand technique. Local anesthesia was routinely applied during procedures. Samples obtained were expelled on glass slides, smeared, and immediately placed in 95% alcohol for Papanicolaou staining. Cytologic reports were divided into the following 5 categories: (1) malignancy, (2) suspicious for malignancy, (3) benign, (4) indeterminate, and (5) inadequate for diagnosis.25 J Ultrasound Med 2014; 33:585–595
Table 1. Basic Characteristics and Sonographic Features for the Patients With Thyroid Nodules Characteristic
Benign
Malignant
Patients (n =192) Male/female 42/84 9/57 Age, y 49 ±12 (16–67) 47 ± 11 (27–72) Single nodule/multiple 54/72 24/42 nodules Nodules (n = 219) 142 77 Diameter, cm 2.2 ± 1.1 (0.7–6.0) 1.4 ± 0.9 (0.5–4.9) Size distribution Diameter ≤1.0 cm 19 35 Diameter 1.1–2.0 cm 49 31 Diameter 2.1–3.0 cm 45 4 Diameter >3.0 cm 29 7 Location Left lobe 72 33 Right lobe 66 39 Isthmus 4 5 Echogenicity Hyperechoic 1 1 Isoechoic 46 1 Hypoechoic 72 75 Mix solid and cystic 23 0 Calcifications None 123 26 Microcalcifications 7 44 Macrocalcifications 12 7 Shape Oval to round 131 45 Taller than wide 2 17 Irregular 9 15 Margin Well defined 112 36 Poorly defined 30 41 Halo sign Present 80 3 Absent 62 74
P .003 .238 .355
ORIGINAL RESEARCH
Virtual Touch Tissue Imaging on Acoustic Radiation Force Impulse Elastography A New Technique for Differential Diagnosis Between Benign and Malignant Thyroid Nodules Yi-Feng Zhang, MD, Yong He, MD, Hui-Xiong Xu, MD, PhD, Xiao-Hong Xu, MD, Chang Liu, MD, Le-Hang Guo, MD, Lin-Na Liu, MD, Jun-Mei Xu, MD Objectives—Acoustic radiation force impulse elastography is a newly developed ultrasound elasticity imaging technique that included both Virtual Touch tissue quantification and Virtual Touch tissue imaging (VTI; Siemens Medical Solutions, Mountain View, CA). This study aimed to evaluate the usefulness of VTI in differentiating malignant from benign thyroid nodules. Methods—This study included 192 consecutive patients with thyroid nodules (n = 219) who underwent surgery for compressive symptoms or suspicion of malignancy. Tissue stiffness on VTI elastography was scored from 1 (soft) to 6 (hard). The VTI scores between malignant and benign thyroid nodules were compared. The intraobserver and interobserver agreement for VTI elastography was also assessed. Received March 15, 2013, from the Department of Medical Ultrasound, Shanghai Tenth People’s Hospital, Tenth People’s Hospital of Tongji University, Shanghai, China (Y.-F.Z., Y.H., H.X.X., C.L., L.-H.G., L.-N.L., J.-M.X.); Department of Ultrasound, First College of Clinical Medical Sciences, China Three Gorges University, Yichang, China (Y.H.); and Department of Ultrasound, Affiliated Hospital of Guangdong Medical College, Zhanjiang, China (X.-H.X.). Revision requested April 25, 2013. Revised manuscript accepted for publication July 29, 2013. This work was supported in part by grant NCET-06-0723 from the Chinese Ministry of Education and the Shanghai Talent Development Project from the Shanghai Human Resources and Social Security Bureau. Drs Zhang and He contributed equally to the manuscript. Address correspondence to Hui-Xiong Xu, MD, PhD, Department of Medical Ultrasound, Shanghai Tenth People’s Hospital , Tenth People's Hospital of Tongji University, 301 Yanchangzhong Rd, 200072 Shanghai, China. E-mail: [email protected] Abbreviations
ARFI, acoustic radiation force impulse; Az , area under the curve; FOV, field of view; NPV, negative predictive value; PPV, positive predictive value; VTI, Virtual Touch tissue imaging; VTQ, Virtual Touch tissue quantification doi:10.7863/ultra.33.4.585
Results—On VTI elastography: score 1 was found in 84 nodules (all benign); score 2 in 37 nodules (3 papillary carcinomas and 34 benign nodules); score 3 in 25 nodules (1 medullary carcinoma, 6 papillary carcinomas, and 18 benign nodules); score 4 in 53 nodules (50 papillary carcinomas and 3 benign nodules); score 5 in 17 nodules (14 papillary carcinomas and 3 benign nodules); and score 6 in 3 nodules (all papillary carcinomas). A VTI elasticity score of 4 or greater was highly predictive of malignancy (P < .01), and the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 87.0% (67 of 77), 95.8% (136 of 142), 91.8% (67 of 73), 93.1% (136 of 146), and 92.7% (203 of 219), respectively. The κ values were 0.69 for intraobserver agreement and 0.85 for interobserver agreement. Conclusions—Virtual Touch tissue elasticity imaging has great potential as an adjunctive tool combined with conventional sonography for differential diagnosis between benign and malignant thyroid nodules. Key Words—acoustic radiation force impulse; superficial structures; thyroid nodule; ultrasound elastography; Virtual Touch tissue imaging
T
hyroid cancer is the most common type of endocrinerelated cancer, and its incidence has risen 2.4-fold over the last 30 years.1–3 The discrimination between malignant and benign thyroid nodules is of paramount important, since the management and prognosis for them varies greatly.4–6 Palpation is the oldest and most frequently used screening method for detecting thyroid gland tumors. In general, almost 5% of the adult population has a palpable thyroid gland nodule.7 The palpation becomes very difficult for clinicians when the nodule is small or located deeply in the thyroid tissue. As a noninvasive, inexpensive, accurate, and repeatable modality, high-frequency (7.5–13 MHz) sonography has been regarded as the first-line imaging modality for thyroid disease. With conven-
©2014 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2014; 33:585–595 | 0278-4297 | www.aium.org
3304jum557-712online_Layout 1 3/19/14 11:30 AM Page 586
Zhang et al—Virtual Touch Tissue Imaging of Thyroid Nodules
tional sonography, the following sonographic features are risk factors for malignancy of a thyroid nodule: solitary nodule, hypoechoic nodule, blurred margin, “taller-thanwide” shape, and microcalcifications.8,9 However, some authors have argued that the echogenicity, shape, border, and internal vascularity did not show any significant differences between benign and malignant nodules.10 To further enhance the diagnostic ability of sonography, ultrasound elastography, including static and real-time elastography, has been introduced into the clinic with an aim of improving the diagnosis of thyroid cancer. Real-time elastography is one of the most widely used elastographic techniques, in which calculation of the tissue elasticity distribution is performed in real time, and the examination results are represented as color-coded images over the conventional B-mode images. The increase in tissue stiffness has been proven to be associated with a higher risk of malignancy.11,12 As for real-time elastography, the sensitivity and specificity for the diagnosis of malignant thyroid nodules were reported to be 86% to 92% and 89% to 97%, respectively.13–17 However, because of the characteristics of external mechanical pressure, individual differences may be present.18 Acoustic radiation force impulse (ARFI) is a newly introduced ultrasound-based elasticity imaging technique that has been applied to the thyroid in recent years.19–24 In ARFI imaging, tissue is mechanically excited by using short-duration acoustic pulses with a fixed transmit frequency to generate localized tissue displacement. In general, ARFI is between 80 and 200 microseconds. The mechanical index value of the ARFI region of interest in the Virtual Touch tissue imaging (VTI; Siemens Medical Solutions, Mountain View, CA) mode reads in the range of 1.0 to 1.6 when the region of interest is placed at a depth range of 0.5 to 8 cm. The displacement results in shear wave propagation away from the region of excitation, and the shear wave speed of the tissue is proportional to the square root of tissue elasticity. By calculating the shear wave velocity, quantitative evaluation of the tissue stiffness is available, which is named virtual touch tissue quantification (VTQ; Siemens Medical Solutions). On the other hand, the strain change under the push pulse is named VTI, which uses an algorithm that is similar to that of conventional ultrasound elastography and is displayed as a gray scale image. A VTI image is a grayscale display of relative tissue stiffness in a user-defined region of interest. This information is computed by examining the displacements of tissue elements in response to an acoustic push pulse. When a stress (longitudinal force) is applied to the tissues in the region of interest, they may behave differently, with some having greater deformation than others. By compar-
586
ing the baseline and deformed information, individual tissue elements may be labeled by their relative stiffness. A bright shade indicates relatively soft (elastic) tissue, whereas a darker shade indicates relatively stiff (nonelastic) tissue. Therefore, VTI is a qualitative mapping of ARFI, and VTQ is a quantitative estimate of tissue stiffness. Until now, VTQ has been used for the diagnosis of thyroid nodules, and the range of specificity for the diagnosis of malignant thyroid nodules was 82.2% to 95%,19–24 whereas the diagnostic value of VTI has not been well established. In this study, 219 thyroid nodules in 192 patients were assessed with VTI elastography with an aim of evaluating the usefulness of VTI in the diagnosis of thyroid nodules.
Materials and Methods Patients From April 2011 to September 2012, 192 consecutive patients with thyroid nodules were included in this study. There were 141 female and 51 male patients with a mean age ± SD of 49 ± 12 years (range, 16–72 years). Seventyeight patients had a single nodule, and 114 had multiple nodules. All of the patients underwent sonographic and VTI examinations, and the nature of all of the target nodules was pathologically confirmed after surgery. The inclusion criteria of the target nodules were as follows: (1) nodule diameter greater than 0.5 cm; (2) solid or almost solid (≤25% cystic) nodules; (3) no treatment performed on the nodules; (4) enough thyroid tissue surrounding the nodule at the same depth; and (5) thyroid surgery performed after sonography. Nodules showing obvious sonographic features of benignity, such as cystic lesions and spongiform echogenicity, were excluded from analysis. For the patients with multiple nodules, the nodules that were suspicious of malignancy (based on conventional sonographic findings) or the largest solid ones were selected for analysis. Finally, 1 nodule in each patient from the 78 patients with a single nodule (including 54 benign and 24 malignant nodules), 1 nodule in each from 89 patients with multiple nodules (including 58 benign and 31 malignant nodules), 2 nodules in each from 23 patients with multiple nodules (including 12 with 2 benign nodules in each, 10 with 2 malignant nodules in each, and 1 with 1 benign and 1 malignant nodule), and 3 nodules in each from 2 patients with multiple nodules (including 1 with 3 benign nodules, and 1 with 1 malignant and 2 benign nodules) were included in this study (Figure 1); thus, in total, 219 thyroid nodules were observed, and the sonographic characteristics are presented in Table 1. The diameter of the nodules ranged from 0.5 to 6.0 cm (mean, 1.9 ± 1.1 cm).
J Ultrasound Med 2014; 33:585–595
3304jum557-712online_Layout 1 3/19/14 11:30 AM Page 587
Zhang et al—Virtual Touch Tissue Imaging of Thyroid Nodules
Figure 1. Patients with thyroid nodules included in the study; b indicates benign; m, malignant, n, number of nodules; and pt, patient.
All of the nodules in this study underwent surgery. The indications for thyroidectomy were as follows: (1) confirmed malignancy by fine-needle aspiration (n = 4); (2) high suspicion of malignancy by sonography or fineneedle aspiration (n = 93); and (3) compressive symptoms caused by the large nodules or associated large nodules (n = 122). In the patients with compressive symptoms, 74 large nodules causing compressive symptoms were included for VTI analysis and were finally removed. For the remaining 48 small nodules, they were removed together with the associated large nodules when the nodules were present in the same lobe. These 48 nodules were included for VTI analysis because they were solid or hypoechoic on sonography, whereas the associated large nodules were not included for VTI analysis because of obvious sonographic features for benignity, such as cysts and a spongiform echo texture (Figure 2). The criteria for fine-needle aspiration were as follows: (1) patients with a high-risk history and nodule size larger than 5 mm; (2) microcalcifications present in nodules and nodule size of 10 mm or larger; (3) solid nodule and hypoechoic nodules and nodule size larger than 10 mm; (4) solid and isoechoic or hyperechoic nodules and nodule size of 10 to 15 mm or larger; and (5) mixed cystic-solid nodules with any suspicious sonographic features and nodule size of 15–20 mm or larger. Fine-needle aspiration was performed with a 22-gauge needle attached to a 5-mL disposable syringe by a freehand technique. Local anesthesia was routinely applied during procedures. Samples obtained were expelled on glass slides, smeared, and immediately placed in 95% alcohol for Papanicolaou staining. Cytologic reports were divided into the following 5 categories: (1) malignancy, (2) suspicious for malignancy, (3) benign, (4) indeterminate, and (5) inadequate for diagnosis.25 J Ultrasound Med 2014; 33:585–595
Table 1. Basic Characteristics and Sonographic Features for the Patients With Thyroid Nodules Characteristic
Benign
Malignant
Patients (n =192) Male/female 42/84 9/57 Age, y 49 ±12 (16–67) 47 ± 11 (27–72) Single nodule/multiple 54/72 24/42 nodules Nodules (n = 219) 142 77 Diameter, cm 2.2 ± 1.1 (0.7–6.0) 1.4 ± 0.9 (0.5–4.9) Size distribution Diameter ≤1.0 cm 19 35 Diameter 1.1–2.0 cm 49 31 Diameter 2.1–3.0 cm 45 4 Diameter >3.0 cm 29 7 Location Left lobe 72 33 Right lobe 66 39 Isthmus 4 5 Echogenicity Hyperechoic 1 1 Isoechoic 46 1 Hypoechoic 72 75 Mix solid and cystic 23 0 Calcifications None 123 26 Microcalcifications 7 44 Macrocalcifications 12 7 Shape Oval to round 131 45 Taller than wide 2 17 Irregular 9 15 Margin Well defined 112 36 Poorly defined 30 41 Halo sign Present 80 3 Absent 62 74
P .003 .238 .355
