- Email: [email protected]
Value of modern sonography in the assessment of meniscal lesions
European Journal of Radiology 81 (2012) 2366–2369
Contents lists available at SciVerse ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Value of modern sonography in the assessment of meniscal lesions Pawel Wareluk a,1 , Kazimierz T. Szopinski b,∗ a b
Department of Diagnostic Imaging, Second Faculty of Medicine, Medical University of Warsaw, ul. Kondratowicza 8, 03-242 Warsaw, Poland Department of Dental and Maxillofacial Radiology, First Faculty of Medicine, Medical University of Warsaw, ul. Nowogrodzka 59, 02-006 Warsaw, Poland
a r t i c l e
i n f o
Article history: Received 14 July 2011 Received in revised form 14 September 2011 Accepted 17 September 2011 Keywords: Ultrasonography Knee Meniscus Meniscal injury
a b s t r a c t The aim of this prospective study was to assess the accuracy of modern ultrasonography in diagnostic imaging of meniscal tears. One hundred and sixty menisci were evaluated in 80 patients (42 females, 38 males, mean age = 36.2 years, range = 16–70 years). Inclusion criteria for the study were twofold: clinical suspicion of meniscal injury and clinical indication for arthroscopy. Knee examination was performed with the Voluson 730 Expert ultrasound system (General Electric). After sonographic examination, all patients underwent arthroscopic procedures within 1–4 days. The final diagnosis of meniscal tears was taken from surgical reports. The overall sensitivity, specificity, positive predictive value and negative predictive value of sonographic examination in the assessment of meniscal tears amounted to 85.4%, 85.7%, 67.3% and 94.4%, respectively. The statistical parameters were not statistically different in medial and lateral menisci. Age, sex, body mass index (BMI), weight, physical activity, mechanism on injury, and time lapse from injury did not have a statistically significant impact on the usefulness of ultrasonography. The highest sensitivity (>90%) was obtained in medial menisci and in patients with a BMI > 25. The highest specificity (>90%) was obtained in lateral menisci, in patients after twisting injuries, in sports injuries, and in recent injuries (time lapse from the injury <1 month). The positive predictive value (PPV) of sonographic examination was higher than 90% only in recent injuries (<1 month), however, the negative predictive value of ultrasound is high, being less than 90% in males with lesions of lateral menisci and in sequelae of sports injuries. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Meniscal tears are the most common injury of soft tissues [1]. The incidence of meniscal injuries reported in 1985 in the United States of America amounted to 61/100,000/year, with a clear male predominance (3:1) [2]. The incidence reported in Great Britain in 2008 amounts to 23 cases per 100,000/year [1]. Meniscal injury may occur due to sports trauma or from a trivial injury due to increased age and a degenerated meniscus. [3]. Although magnetic resonance imaging is currently the diagnostic method of choice in assessment of menisci [4], it is expensive, has a number of contraindications, and is not always readily accessible. Ultrasound has been used for more than three decades in the assessment of the musculoskeletal system, and is the method of choice in the assessment of superficial structures, such as muscles, tendons, and nerves. However, its use in the assessment of menisci remains controversial, despite development of new
∗ Corresponding author. Tel.: +48 225021272; fax: +48 225022112. E-mail addresses: [email protected] (P. Wareluk), [email protected], [email protected] (K.T. Szopinski). 1 Tel.: +48 22 326 5810; fax: +48 22 326 5991. 0720-048X/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2011.09.013
sonographic techniques and dramatic improvements in the quality of sonographic images.
2. Objectives The objectives of this study are: to evaluate the usefulness and value of modern ultrasonography in the assessment of meniscal injuries and to assess the impact of various characteristics of the injury (etiology of injury, sports activity, time elapsed from injury) and the patient (specifically, sex, age, weight, and BMI).
3. Material Sonographic examinations of 85 knee joints in 85 patients (45 females, 40 males, mean age: 36.2 years, range 16–70 years) were performed in a two-year period from 2005 to 2007. The inclusion criteria for the study were: clinical suspicion of meniscal injury as stated by an orthopedic surgeon, and clinical indication for arthroscopy. Five patients, who despite clinical indication for arthroscopic surgery, did not undergo it and were excluded from the study. The final study population comprised 80 patients (42 females, 38 males). The etiology (mechanism) of injury, time
P. Wareluk, K.T. Szopinski / European Journal of Radiology 81 (2012) 2366–2369
2367
elapsed from injury, sex, age, weight, BMI and physical activity of the patients were recorded. Informed consent was obtained from all patients. Since sonographic knee examination is part of normal medical procedure for all patients referred for arthroscopic surgery with suspicion of meniscal tear, the study was waived from institutional board approval since it did not incur any additional health hazards for the patient nor did it breach patient confidentiality. 4. Methods All sonographic examinations were performed by a single physician with 7 years experience in musculoskeletal ultrasound. Examinations were performed with the Voluson 730 Expert ultrasound system (General Electric), using a 6–12 MHz frequency probe, to allow automatic volumetric data acquisition. A variety of advanced techniques improving image quality were used (compound imaging, speckle reduction, and harmonic imaging). In each knee, both menisci were evaluated. The knees were examined in supine and prone positions in full extension and 30◦ flexion. Dynamic examinations aimed at showing displacement of the fragments of the meniscus were also performed during slight passive varus-valgus movements. Following sonographic examination, criteria for the diagnosis of meniscal tear included: hypoechoic line extending to the surface of the meniscus, irregular outline, lacking sharp edge or presence of a meniscal cyst (Figs. 1–3). All patients underwent knee arthroscopy within 1–4 days after sonographic examination. The surgical reports were reviewed and compared with the results of sonographic examinations. 5. Statistical evaluation To compare the results of ultrasonography and arthroscopy chisquare test and Fisher exact test were used. Statistical significance level was set at p < 0.05. Sensitivity, specificity, positive predictive
Fig. 1. A bucket-handle tear of the medial meniscus demonstrated on twodimensional ultrasonography.
value, and negative predictive value of sonographic examination were calculated. 6. Results A correct sonographic diagnosis was obtained in 35 (85,4%) meniscal tears and 102 (85.7%) intact menisci as seen on arthroscopy. The concordance of sonographic and arthroscopic diagnosis was statistically significant (p < 0.05). Detailed results are shown in Table 1. The overall sensitivity, specificity, positive predictive value, and negative predictive value of sonographic examination in the assessment of meniscal tears amounted to 85.4%, 85.7%, 67.5% and 94.4%, respectively. The results obtained in meniscal tears secondary to twisting injuries (sensitivity = 91.3%) were better than in meniscal
Fig. 2. A bucket-handle tear of the medial meniscus demonstrated on three-dimensional ultrasonography (multi-planar reformation).
2368
P. Wareluk, K.T. Szopinski / European Journal of Radiology 81 (2012) 2366–2369
Table 1 Sensitivity, specificity, positive predictive value (PPV) and negative predictive value of sonography in assessment of meniscal lesions.
All menisci Lateral menisci Medial menisci Menisci in female patients Menisci in male patients Lateral menisci in females Lateral menisci in males Medial menisci in females Medical menisci in males Age < 30 Age 30–50 Age > 50 Twisting injury Blunt injury Time from injury <1 month Time from injury 1–6 months Time from injury >6 months BMI < 25 BMI > 25 Sports injuries Non-sports injuries
n
Sensitivity (95% CI)
Specificity (95% CI)
PPV
NPV
160 80 80 84 76 42 38 42 38 70 50 40 80 36 32 22 64 72 88 64 96
0.854 (0.745–0.962) 0.667 (0.400–0.933) 0.931 (0.839–1.023) 0.857 (0.674–1.040) 0.852 (0.718–0.986) 0.750 (0.326–1.174) 0.625 (0.290–0.960) 0.900 (0.714–1.086) 0.947 (0.847–1.048) 0.813 (0.621–1.004) 0.867 (0.695–1.039) 0.900 (0.714–1.086) 0.857 (0.707–1.007) 0.778 (0.506–1.049) 0.833 (0.622–1.044) 0.857 (0.598–1.116) 0.833 (0.622–1.044) 0.789 (0.606–0.973) 0.909 (0.789–1.029) 0.722 (0.515–0.929) 0.957 (0.873–1.040)
0.857 (0.794–0.920) 0.956 (0.907–1.055) 0.725 (0.603–0.848) 0.829 (0.740–0.917) 0.898 (0.813–0.983) 0.947 (0.876–1.018) 0.952 (0.861–1.043) 0.688 (0.527–0.848) 0.789 (0.606–0.973) 0.870 (0.781–0.960) 0.914 (0.822–1.007) 0.767 (0.615–0.918) 0.932 (0.868–0.996) 0.704 (0.513–0.876) 0.950 (0.854–1.046) 0.800 (0.509–1.002) 0.846 (0.748–0.944) 0.868 (0.777–0.959) 0.848 (0.762–0.935) 0.913 (0.832–0.994) 0.822 (0.734–0.910)
0.673 0.727 0.659 0.500 0.821 0.600 0.833 0.474 0.818 0.650 0.813 0.563 0.818 0.467 0.909 0.667 0.556 0.682 0.667 0.765 0.629
0.944 0.942 0.949 0.967 0.917 0.973 0.870 0.957 0.938 0.940 0.941 0.958 0.948 0.905 0.905 0.923 0.957 0.920 0.966 0.894 0.984
tears secondary to blunt injuries (sensitivity = 72.2%). A possible explanation for this result is that more extensive damage to the meniscus (caused by stronger external forces in cases of twisting injury) is observed more easily. Better results were also achieved in cases of recent injuries than in older ones but the differences were not statistically significant. Sex of the patients, their weight (BMI), and physical activity did not have a significant impact on the results. The results obtained in medial and lateral menisci did not differ significantly. Due to incomplete data from surgical procedures (in some cases the type of meniscal injury was not described) the differentiation of meniscal tear types was not possible in the majority of cases. In 59 patients (118 menisci) the time lapse between injury and sonographic examination could be determined. The patients were subdivided into 3 groups: less than 1 month from injury (16 patients), 1–6 months from injury (11 patients), and more than 6 months from injury (32 patients).
Fig. 3. A horizontal tear of the posterior horn of the medial meniscus demonstrated on two-dimensional ultrasonography.
7. Discussion Sonography has been used in the diagnosis of meniscal tears for more than 20 years [5]. Surprisingly, despite major advances in sonographic technology, an extremely wide range of sensitivity and specificity values of this method have been reported. Some authors advocate use of ultrasound in the assessment of menisci [6], while others state that this technique should not be used [7]. Since orthopedic surgeons often refer patients with suspicion of meniscal lesions for knee ultrasound in our practice, we wanted to assess the value of this technique. Ultrasound is an attractive imaging modality in the assessment of knee trauma as it is safe, inexpensive, and easily accessible. Small portable machines can aid in the primary assessment of the injury on the field. The main limitation is its inherent inability to depict deep structures of the knee, specifically, the anterior cruciate ligament and knee cartilages. Moreover, the sensitivity and specificity of sonographic assessment of menisci remains a controversial issue. The reported sensitivity and specificity of ultrasound in diagnosis of meniscal tears varies widely – from 0% to 97.2% and from 21% to 100% respectively [8–12]. Although a detailed meta-analysis is beyond the scope of this paper, it is worth noting that the results obtained at present and 20 years ago largely overlap, and the diagnostic performance of musculoskeletal radiologists and orthopedic surgeons is similar. The lowest reported sensitivity values were obtained when examinations were performed in general radiology practices [12]. Our results are close to those obtained by Riedl et al. [13]. Although the overall sensitivity is reasonable (85.4%), it varies largely in specific sub-groups (from 62.5% in the lateral menisci in males to 95.7% in non-sports injuries). The overall specificity is also reasonable (85.7%), however with less variability (from 70.4% in sequelae of blunt injuries to 95.6% in detection of tears of the lateral menisci). The overall positive predictive value is relatively poor (67.3%, varying from 46.7% in the sequelae of blunt trauma to 90.9% in recent trauma). However, the negative predictive value of sonography was uniformly good (overall 94.4%, varying from 87% in the lateral menisci in males to 98.4% in non-sports injuries). Therefore, ultrasound should not be used as a sole or principal method of assessment of the menisci. We agree with the opinion of the American Institute of Ultrasound in Medicine (AIUM) guidelines that the assessment of menisci should be performed in
P. Wareluk, K.T. Szopinski / European Journal of Radiology 81 (2012) 2366–2369
all cases of sonographic examination of the knee and if meniscal pathology is suspected, either clinically or by ultrasound, further examination with magnetic resonance imaging should be performed [6]. However, our results show that a negative result of sonographic examination rules out meniscal damage with high probability. The sensitivity of ultrasound was higher in diagnosis of medial meniscal tears, specificity was better in lateral menisci, which is in agreement with the results of Khan et al. [9] and Helwig et al. [14,15]. The best overall results (defined as the sum of sensitivity, specificity, PPV and NPV values) were obtained in recent injuries (less than 1 month before examination). The worst results were obtained in cases of chronic, blunt damage to the meniscus. A large number of false positives in female patients resulted in poor positive predictive values of ultrasound in this group. Presence of degenerative changes mimicking meniscal tears seems to be the most plausible explanation. To our surprise, being overweight (as defined by a BMI > 25) did not significantly affect the diagnostic performance of meniscal ultrasound. This study has several limitations. In a large proportion of cases the extent and/or location of the tear was not noted in the surgical reports available to us, so no conclusions on the value of the ultrasound in detection of meniscal tears in specific locations could be drawn. We did not analyze the impact of separate techniques improving the quality of sonographic imaging, since the study was aimed at studying ultrasonography as a whole. 8. Conclusions Modern sonography has a high negative predictive value and a reasonable specificity in the assessment of meniscal injuries of the knee. It is particularly useful in the assessment of recent meniscal injuries. The diagnostic performance of sonography did not significantly differ in medial and lateral menisci. The usefulness of ultrasonography was not significantly affected by other factors examined in the study including sex, age, weight (BMI), or physical activity of the patients.
2369
Conflict of interest statement The authors or the authors’ institution have no conflicts of interest. Acknowledgements We would gratefully acknowledge the help of Professor Jaroslaw Deszczynski, Head of the Clinic of Orthopedic Surgery and Rehabilitation, Warsaw Medical University for making available to us the results of arthroscopic procedures, and Natalia Ziolkowski, B.Com, B.Sc for reviewing the manuscript. References [1] Clayton RAE, Court-Brown CM. The epidemiology of musculoskeletal tendinous and ligamentous injuries. Injury 2008;39(12):1338–44. [2] Baker BE, Peckham AC, Pupparo F, Sanborn JC. Review of meniscal injury and associated sports. Am J Sports Med 1985;13(1):1–4. [3] Rath E, Richmond JC. The menisci: basic science and advancements in treatment. Br J Sports Med 2000;34:252–7. [4] Oei EHG, Ginai AZ, Hunink MGM. MRI for traumatic knee injury – a review. Semin Ultrasound CT MRI 2007;28:141–57. [5] Heuchemer T, Bauer G, Friedrich JM, Bargon G. Clinical use of meniscus sonography. Bildgebung 1987–1989;56(3):118–23. [6] http://www.aium.org/publications/guidelines/musculoskeletal.pdf [accessed on 19.06.11]. [7] Martinoli C, Bianchi S. Knee. In: Bianchi S, Martinoli C, editors. Ultrasound of the musculoskeletal system. Berlin Heidelberg: Springer-Verlag; 2007. [8] Azzoni R, Cabitza P. Is there a role for ultrasonography in the diagnosis of tears of the knee menisci? J Clin Ultrasound 2002;30(8):472–6. [9] Khan Z, Faruqui Z, Ogbunbiyi O, Rosset G, Iqbal J. Ultrasound assessment of internal derangement of the knee. Acta Orthop Belg 2006;72(1):72–6. [10] Shanbhogue AKP, Sandhu MS, Singh P, Ojili V, Khandeval N, Sen R. Real time spatial compound ultrasound in the evaluation of meniscal injuries – a comparison study with conventional ultrasound and MRI. Knee 2009;16:191–5. [11] Timotijevic´ S, Vukaˇsinovic´ Z, Baˇscˇ arevic´ Z. Validity of clinical and ultrasound examination related to arthroscopy in acute injury of the medial meniscus of the knee. Srp Arh Celok Lek 2008;136(1–2):28–32. [12] Warwick B, Lee TS, Sundarajan V, Walker P, Magnussen J, van der Wall H. Performance characteristics of ultrasound of the knee in a general radiological setting. Knee 2004;11:303–6. [13] Riedl S, Tauscher A, Kühner C, Göhring U, Sohn C, Meeder PJ. Die dreidimensionale Sonographie in der Diagnostik der Meniskusläsionen. Chirurg 1997;68:1150–5. [14] Helwig P, Hauschild O, Bahrs C, Weise K, Schewe B. 3-Dimensional ultrasound imaging for meniscal lesions. Knee 2007;14:478–83. [15] Helwig P, Bahrs C, Weise K, Schewe B. Wertigkeit von 3D-Sonographie und MRT des Meniskusschadens. Orthopäde 2006;35:982–8.
Contents lists available at SciVerse ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Value of modern sonography in the assessment of meniscal lesions Pawel Wareluk a,1 , Kazimierz T. Szopinski b,∗ a b
Department of Diagnostic Imaging, Second Faculty of Medicine, Medical University of Warsaw, ul. Kondratowicza 8, 03-242 Warsaw, Poland Department of Dental and Maxillofacial Radiology, First Faculty of Medicine, Medical University of Warsaw, ul. Nowogrodzka 59, 02-006 Warsaw, Poland
a r t i c l e
i n f o
Article history: Received 14 July 2011 Received in revised form 14 September 2011 Accepted 17 September 2011 Keywords: Ultrasonography Knee Meniscus Meniscal injury
a b s t r a c t The aim of this prospective study was to assess the accuracy of modern ultrasonography in diagnostic imaging of meniscal tears. One hundred and sixty menisci were evaluated in 80 patients (42 females, 38 males, mean age = 36.2 years, range = 16–70 years). Inclusion criteria for the study were twofold: clinical suspicion of meniscal injury and clinical indication for arthroscopy. Knee examination was performed with the Voluson 730 Expert ultrasound system (General Electric). After sonographic examination, all patients underwent arthroscopic procedures within 1–4 days. The final diagnosis of meniscal tears was taken from surgical reports. The overall sensitivity, specificity, positive predictive value and negative predictive value of sonographic examination in the assessment of meniscal tears amounted to 85.4%, 85.7%, 67.3% and 94.4%, respectively. The statistical parameters were not statistically different in medial and lateral menisci. Age, sex, body mass index (BMI), weight, physical activity, mechanism on injury, and time lapse from injury did not have a statistically significant impact on the usefulness of ultrasonography. The highest sensitivity (>90%) was obtained in medial menisci and in patients with a BMI > 25. The highest specificity (>90%) was obtained in lateral menisci, in patients after twisting injuries, in sports injuries, and in recent injuries (time lapse from the injury <1 month). The positive predictive value (PPV) of sonographic examination was higher than 90% only in recent injuries (<1 month), however, the negative predictive value of ultrasound is high, being less than 90% in males with lesions of lateral menisci and in sequelae of sports injuries. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Meniscal tears are the most common injury of soft tissues [1]. The incidence of meniscal injuries reported in 1985 in the United States of America amounted to 61/100,000/year, with a clear male predominance (3:1) [2]. The incidence reported in Great Britain in 2008 amounts to 23 cases per 100,000/year [1]. Meniscal injury may occur due to sports trauma or from a trivial injury due to increased age and a degenerated meniscus. [3]. Although magnetic resonance imaging is currently the diagnostic method of choice in assessment of menisci [4], it is expensive, has a number of contraindications, and is not always readily accessible. Ultrasound has been used for more than three decades in the assessment of the musculoskeletal system, and is the method of choice in the assessment of superficial structures, such as muscles, tendons, and nerves. However, its use in the assessment of menisci remains controversial, despite development of new
∗ Corresponding author. Tel.: +48 225021272; fax: +48 225022112. E-mail addresses: [email protected] (P. Wareluk), [email protected], [email protected] (K.T. Szopinski). 1 Tel.: +48 22 326 5810; fax: +48 22 326 5991. 0720-048X/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2011.09.013
sonographic techniques and dramatic improvements in the quality of sonographic images.
2. Objectives The objectives of this study are: to evaluate the usefulness and value of modern ultrasonography in the assessment of meniscal injuries and to assess the impact of various characteristics of the injury (etiology of injury, sports activity, time elapsed from injury) and the patient (specifically, sex, age, weight, and BMI).
3. Material Sonographic examinations of 85 knee joints in 85 patients (45 females, 40 males, mean age: 36.2 years, range 16–70 years) were performed in a two-year period from 2005 to 2007. The inclusion criteria for the study were: clinical suspicion of meniscal injury as stated by an orthopedic surgeon, and clinical indication for arthroscopy. Five patients, who despite clinical indication for arthroscopic surgery, did not undergo it and were excluded from the study. The final study population comprised 80 patients (42 females, 38 males). The etiology (mechanism) of injury, time
P. Wareluk, K.T. Szopinski / European Journal of Radiology 81 (2012) 2366–2369
2367
elapsed from injury, sex, age, weight, BMI and physical activity of the patients were recorded. Informed consent was obtained from all patients. Since sonographic knee examination is part of normal medical procedure for all patients referred for arthroscopic surgery with suspicion of meniscal tear, the study was waived from institutional board approval since it did not incur any additional health hazards for the patient nor did it breach patient confidentiality. 4. Methods All sonographic examinations were performed by a single physician with 7 years experience in musculoskeletal ultrasound. Examinations were performed with the Voluson 730 Expert ultrasound system (General Electric), using a 6–12 MHz frequency probe, to allow automatic volumetric data acquisition. A variety of advanced techniques improving image quality were used (compound imaging, speckle reduction, and harmonic imaging). In each knee, both menisci were evaluated. The knees were examined in supine and prone positions in full extension and 30◦ flexion. Dynamic examinations aimed at showing displacement of the fragments of the meniscus were also performed during slight passive varus-valgus movements. Following sonographic examination, criteria for the diagnosis of meniscal tear included: hypoechoic line extending to the surface of the meniscus, irregular outline, lacking sharp edge or presence of a meniscal cyst (Figs. 1–3). All patients underwent knee arthroscopy within 1–4 days after sonographic examination. The surgical reports were reviewed and compared with the results of sonographic examinations. 5. Statistical evaluation To compare the results of ultrasonography and arthroscopy chisquare test and Fisher exact test were used. Statistical significance level was set at p < 0.05. Sensitivity, specificity, positive predictive
Fig. 1. A bucket-handle tear of the medial meniscus demonstrated on twodimensional ultrasonography.
value, and negative predictive value of sonographic examination were calculated. 6. Results A correct sonographic diagnosis was obtained in 35 (85,4%) meniscal tears and 102 (85.7%) intact menisci as seen on arthroscopy. The concordance of sonographic and arthroscopic diagnosis was statistically significant (p < 0.05). Detailed results are shown in Table 1. The overall sensitivity, specificity, positive predictive value, and negative predictive value of sonographic examination in the assessment of meniscal tears amounted to 85.4%, 85.7%, 67.5% and 94.4%, respectively. The results obtained in meniscal tears secondary to twisting injuries (sensitivity = 91.3%) were better than in meniscal
Fig. 2. A bucket-handle tear of the medial meniscus demonstrated on three-dimensional ultrasonography (multi-planar reformation).
2368
P. Wareluk, K.T. Szopinski / European Journal of Radiology 81 (2012) 2366–2369
Table 1 Sensitivity, specificity, positive predictive value (PPV) and negative predictive value of sonography in assessment of meniscal lesions.
All menisci Lateral menisci Medial menisci Menisci in female patients Menisci in male patients Lateral menisci in females Lateral menisci in males Medial menisci in females Medical menisci in males Age < 30 Age 30–50 Age > 50 Twisting injury Blunt injury Time from injury <1 month Time from injury 1–6 months Time from injury >6 months BMI < 25 BMI > 25 Sports injuries Non-sports injuries
n
Sensitivity (95% CI)
Specificity (95% CI)
PPV
NPV
160 80 80 84 76 42 38 42 38 70 50 40 80 36 32 22 64 72 88 64 96
0.854 (0.745–0.962) 0.667 (0.400–0.933) 0.931 (0.839–1.023) 0.857 (0.674–1.040) 0.852 (0.718–0.986) 0.750 (0.326–1.174) 0.625 (0.290–0.960) 0.900 (0.714–1.086) 0.947 (0.847–1.048) 0.813 (0.621–1.004) 0.867 (0.695–1.039) 0.900 (0.714–1.086) 0.857 (0.707–1.007) 0.778 (0.506–1.049) 0.833 (0.622–1.044) 0.857 (0.598–1.116) 0.833 (0.622–1.044) 0.789 (0.606–0.973) 0.909 (0.789–1.029) 0.722 (0.515–0.929) 0.957 (0.873–1.040)
0.857 (0.794–0.920) 0.956 (0.907–1.055) 0.725 (0.603–0.848) 0.829 (0.740–0.917) 0.898 (0.813–0.983) 0.947 (0.876–1.018) 0.952 (0.861–1.043) 0.688 (0.527–0.848) 0.789 (0.606–0.973) 0.870 (0.781–0.960) 0.914 (0.822–1.007) 0.767 (0.615–0.918) 0.932 (0.868–0.996) 0.704 (0.513–0.876) 0.950 (0.854–1.046) 0.800 (0.509–1.002) 0.846 (0.748–0.944) 0.868 (0.777–0.959) 0.848 (0.762–0.935) 0.913 (0.832–0.994) 0.822 (0.734–0.910)
0.673 0.727 0.659 0.500 0.821 0.600 0.833 0.474 0.818 0.650 0.813 0.563 0.818 0.467 0.909 0.667 0.556 0.682 0.667 0.765 0.629
0.944 0.942 0.949 0.967 0.917 0.973 0.870 0.957 0.938 0.940 0.941 0.958 0.948 0.905 0.905 0.923 0.957 0.920 0.966 0.894 0.984
tears secondary to blunt injuries (sensitivity = 72.2%). A possible explanation for this result is that more extensive damage to the meniscus (caused by stronger external forces in cases of twisting injury) is observed more easily. Better results were also achieved in cases of recent injuries than in older ones but the differences were not statistically significant. Sex of the patients, their weight (BMI), and physical activity did not have a significant impact on the results. The results obtained in medial and lateral menisci did not differ significantly. Due to incomplete data from surgical procedures (in some cases the type of meniscal injury was not described) the differentiation of meniscal tear types was not possible in the majority of cases. In 59 patients (118 menisci) the time lapse between injury and sonographic examination could be determined. The patients were subdivided into 3 groups: less than 1 month from injury (16 patients), 1–6 months from injury (11 patients), and more than 6 months from injury (32 patients).
Fig. 3. A horizontal tear of the posterior horn of the medial meniscus demonstrated on two-dimensional ultrasonography.
7. Discussion Sonography has been used in the diagnosis of meniscal tears for more than 20 years [5]. Surprisingly, despite major advances in sonographic technology, an extremely wide range of sensitivity and specificity values of this method have been reported. Some authors advocate use of ultrasound in the assessment of menisci [6], while others state that this technique should not be used [7]. Since orthopedic surgeons often refer patients with suspicion of meniscal lesions for knee ultrasound in our practice, we wanted to assess the value of this technique. Ultrasound is an attractive imaging modality in the assessment of knee trauma as it is safe, inexpensive, and easily accessible. Small portable machines can aid in the primary assessment of the injury on the field. The main limitation is its inherent inability to depict deep structures of the knee, specifically, the anterior cruciate ligament and knee cartilages. Moreover, the sensitivity and specificity of sonographic assessment of menisci remains a controversial issue. The reported sensitivity and specificity of ultrasound in diagnosis of meniscal tears varies widely – from 0% to 97.2% and from 21% to 100% respectively [8–12]. Although a detailed meta-analysis is beyond the scope of this paper, it is worth noting that the results obtained at present and 20 years ago largely overlap, and the diagnostic performance of musculoskeletal radiologists and orthopedic surgeons is similar. The lowest reported sensitivity values were obtained when examinations were performed in general radiology practices [12]. Our results are close to those obtained by Riedl et al. [13]. Although the overall sensitivity is reasonable (85.4%), it varies largely in specific sub-groups (from 62.5% in the lateral menisci in males to 95.7% in non-sports injuries). The overall specificity is also reasonable (85.7%), however with less variability (from 70.4% in sequelae of blunt injuries to 95.6% in detection of tears of the lateral menisci). The overall positive predictive value is relatively poor (67.3%, varying from 46.7% in the sequelae of blunt trauma to 90.9% in recent trauma). However, the negative predictive value of sonography was uniformly good (overall 94.4%, varying from 87% in the lateral menisci in males to 98.4% in non-sports injuries). Therefore, ultrasound should not be used as a sole or principal method of assessment of the menisci. We agree with the opinion of the American Institute of Ultrasound in Medicine (AIUM) guidelines that the assessment of menisci should be performed in
P. Wareluk, K.T. Szopinski / European Journal of Radiology 81 (2012) 2366–2369
all cases of sonographic examination of the knee and if meniscal pathology is suspected, either clinically or by ultrasound, further examination with magnetic resonance imaging should be performed [6]. However, our results show that a negative result of sonographic examination rules out meniscal damage with high probability. The sensitivity of ultrasound was higher in diagnosis of medial meniscal tears, specificity was better in lateral menisci, which is in agreement with the results of Khan et al. [9] and Helwig et al. [14,15]. The best overall results (defined as the sum of sensitivity, specificity, PPV and NPV values) were obtained in recent injuries (less than 1 month before examination). The worst results were obtained in cases of chronic, blunt damage to the meniscus. A large number of false positives in female patients resulted in poor positive predictive values of ultrasound in this group. Presence of degenerative changes mimicking meniscal tears seems to be the most plausible explanation. To our surprise, being overweight (as defined by a BMI > 25) did not significantly affect the diagnostic performance of meniscal ultrasound. This study has several limitations. In a large proportion of cases the extent and/or location of the tear was not noted in the surgical reports available to us, so no conclusions on the value of the ultrasound in detection of meniscal tears in specific locations could be drawn. We did not analyze the impact of separate techniques improving the quality of sonographic imaging, since the study was aimed at studying ultrasonography as a whole. 8. Conclusions Modern sonography has a high negative predictive value and a reasonable specificity in the assessment of meniscal injuries of the knee. It is particularly useful in the assessment of recent meniscal injuries. The diagnostic performance of sonography did not significantly differ in medial and lateral menisci. The usefulness of ultrasonography was not significantly affected by other factors examined in the study including sex, age, weight (BMI), or physical activity of the patients.
2369
Conflict of interest statement The authors or the authors’ institution have no conflicts of interest. Acknowledgements We would gratefully acknowledge the help of Professor Jaroslaw Deszczynski, Head of the Clinic of Orthopedic Surgery and Rehabilitation, Warsaw Medical University for making available to us the results of arthroscopic procedures, and Natalia Ziolkowski, B.Com, B.Sc for reviewing the manuscript. References [1] Clayton RAE, Court-Brown CM. The epidemiology of musculoskeletal tendinous and ligamentous injuries. Injury 2008;39(12):1338–44. [2] Baker BE, Peckham AC, Pupparo F, Sanborn JC. Review of meniscal injury and associated sports. Am J Sports Med 1985;13(1):1–4. [3] Rath E, Richmond JC. The menisci: basic science and advancements in treatment. Br J Sports Med 2000;34:252–7. [4] Oei EHG, Ginai AZ, Hunink MGM. MRI for traumatic knee injury – a review. Semin Ultrasound CT MRI 2007;28:141–57. [5] Heuchemer T, Bauer G, Friedrich JM, Bargon G. Clinical use of meniscus sonography. Bildgebung 1987–1989;56(3):118–23. [6] http://www.aium.org/publications/guidelines/musculoskeletal.pdf [accessed on 19.06.11]. [7] Martinoli C, Bianchi S. Knee. In: Bianchi S, Martinoli C, editors. Ultrasound of the musculoskeletal system. Berlin Heidelberg: Springer-Verlag; 2007. [8] Azzoni R, Cabitza P. Is there a role for ultrasonography in the diagnosis of tears of the knee menisci? J Clin Ultrasound 2002;30(8):472–6. [9] Khan Z, Faruqui Z, Ogbunbiyi O, Rosset G, Iqbal J. Ultrasound assessment of internal derangement of the knee. Acta Orthop Belg 2006;72(1):72–6. [10] Shanbhogue AKP, Sandhu MS, Singh P, Ojili V, Khandeval N, Sen R. Real time spatial compound ultrasound in the evaluation of meniscal injuries – a comparison study with conventional ultrasound and MRI. Knee 2009;16:191–5. [11] Timotijevic´ S, Vukaˇsinovic´ Z, Baˇscˇ arevic´ Z. Validity of clinical and ultrasound examination related to arthroscopy in acute injury of the medial meniscus of the knee. Srp Arh Celok Lek 2008;136(1–2):28–32. [12] Warwick B, Lee TS, Sundarajan V, Walker P, Magnussen J, van der Wall H. Performance characteristics of ultrasound of the knee in a general radiological setting. Knee 2004;11:303–6. [13] Riedl S, Tauscher A, Kühner C, Göhring U, Sohn C, Meeder PJ. Die dreidimensionale Sonographie in der Diagnostik der Meniskusläsionen. Chirurg 1997;68:1150–5. [14] Helwig P, Hauschild O, Bahrs C, Weise K, Schewe B. 3-Dimensional ultrasound imaging for meniscal lesions. Knee 2007;14:478–83. [15] Helwig P, Bahrs C, Weise K, Schewe B. Wertigkeit von 3D-Sonographie und MRT des Meniskusschadens. Orthopäde 2006;35:982–8.