Knees of Ironman Triathletes: Magnetic Resonance Imaging Assessment of Older (>35 Years Old) Competitors

Transcription

1 JOURNAL OF MAGNETIC RESONANCE IMAGING 17: (2003) Original Research Knees of Ironman Triathletes: Magnetic Resonance Imaging Assessment of Older (>35 Years Old) Competitors Frank G. Shellock, PhD, 1,2 * W. Douglas B. Hiller, MD, 3,4 George R. Ainge, MD, 5 David W. Brown, MD, 5 and Laura Dierenfield, AB 4 Purpose: To use magnetic resonance imaging (MRI) to evaluate the knees of older ( 35 years old), competitive Ironman triathletes to determine the prevalence of abnormal findings. Materials and Methods: The knees of 29 Ironman triathletes (20 men, 9 women; age range, years old) were studied by MRI. The findings were analyzed collectively and categorized into group I (N 13), subjects without prior knee injuries and symptoms, and group II (N 16), subjects with prior knee injuries and/or current symptoms. Results: Ten percent of the knees had ligamentous abnormalities, but the prevalence was not statistically different comparing group I to group II. Fifty-five percent had abnormal menisci. The overall prevalence of abnormal menisci was significantly higher in group II (69%) than in group I (38%, P 0.05). Cartilage abnormalities were found in 21% of the triathletes with a higher prevalence in group II (31%) than in group I (8%, P 0.05). Twenty-one percent (6/29) of the knees had bone contusions, with a higher prevalence in group II (31%) than in group I (8%, P 0.05). Conclusion: In general, the spectrum of abnormal MRI findings of the knee was no greater than age-related changes previously reported for other athletic populations and nonathletes. These results have important implications for the diagnostic use of MRI of the knee in this high-endurance, athletic population. 1 Institute for Magnetic Resonance Safety, Education, and Research, Los Angeles, California. 2 University of Southern California, Los Angeles, California. 3 North Hawaii Community Hospital, Kamuela, Hawaii. 4 Labman Hawaii, Kamuela, Hawaii. 5 Department of Radiology, North Hawaii Community Hospital, Kamuela, Hawaii. Contract grant sponsor: North Hawaii Community Hospital, Kamuela, HI; Contract grant sponsor: Labman Hawaii, Inc.; Contract grant sponsor: Institute for Magnetic Resonance Safety, Education, and Research, Los Angeles, CA. This paper was presented as an abstract at the International Society for Magnetic Resonance in Medicine, Annual Meeting, Honolulu, HI, *Address reprint requests to: F.G.S., Institute for Magnetic Resonance, Safety, Education, and Research, 7511 McConnell Ave., Los Angeles, CA frank.shellock@gte.net; Received May 14, 2002; Accepted September 18, DOI /jmri Published online in Wiley InterScience ( Key Words: magnetic resonance imaging (MRI); knee; ligaments; cartilage; bone contusion; musculoskeletal; injuries; athletics J. Magn. Reson. Imaging 2003;17: Wiley-Liss, Inc. CONCERN EXISTS THAT CERTAIN TYPES of exercises, particularly vigorous activity, may increase physical disabilities, especially as athletes get older (1 9). Excessive participation in sports, overtraining, or long-term training can increase the risk of developing osteoarthritis (1 9). Presumably, the continuous stress that prolonged physical activity places on joints can result in microtrauma and degeneration (5). For example, athletes participating in contact sports or activities such as excessive running have a greater risk for the development of osteoarthritis of the knee (6,8). Additionally, older athletes may have a greater prevalence of osteoarthritis in joints, especially those that have suffered injuries (2,4,5,7,9,10). Triathlon competitions involve swimming, bicycling, and running over various distances and under a variety of technical conditions. The Ironman triathlon, comprised of a 2.4-mile swim, 112-mile bike race, and mile run, is one of the best-known triathlon events. Obviously, competing in this high-level, endurance activity requires considerable training by triathletes. For older ( 35 years old) triathlete competitors that have participated in this sport for several years, the associated repetitive musculoskeletal loads may have an adverse, cumulative effect on their joints. In general, the knee is a critical musculoskeletal anatomic site for competitive and recreational athletes. For triathletes, the knee is a frequently injured joint and/or site of painful symptoms (11 14). To date, there has been no research conducted to assess the knees of triathletes to determine the prevalence of meniscal abnormalities, ligament abnormalities, articular cartilage defects, hematopoietic bone marrow hyperplasia, bone contusions, or other conditions. Notably, it is unknown if older triathletes have a greater prevalence of certain conditions than other athletic populations or nonathletic subjects. Additionally, the long-term effects of 2002 Wiley-Liss, Inc. 122

2 Knees of Ironman Triathletes: MRI Assessment 123 training and competing as a triathlete, particularly at the Ironman triathlon level, on the knee have not been determined. Therefore, the goal of this investigation was to use magnetic resonance imaging (MRI) to evaluate the knees of older ( 35 years old), competitive Ironman triathletes to determine the prevalence of abnormal findings. To our knowledge, this is the first such study performed in this unique population of highly trained, endurance athletes. MATERIALS AND METHODS Study Subjects Volunteer study subjects were recruited for this investigation. Inclusion criteria were 1) age 35 years old or older, 2) 4 years or longer participation as an Ironman triathlon competitor, and 3) no prior knee surgery. Exclusion criteria were 1) previous knee surgery or 2) the presence of a condition associated with a contraindication to MRI. Accordingly, the knees of 29 Ironman triathletes (20 men, 9 women; average age, 47 years old; age range, years old) were studied by MRI. All triathletes were interviewed prior to the MRI examination, with particular attention to obtaining training and knee-related information. Data pertaining to their training history and information pertaining to their knees were recorded using a standardized form to facilitate analysis. Because of the known relationships between previous knee injuries or current symptoms and knee abnormalities (2,4,5,7,9,10), the data for the study subjects were analyzed collectively and categorized into two groups, as follows: group I (N 13), study subjects without prior knee injuries and no current symptoms; and group II (N 16), study subjects with prior knee injuries and/or current symptoms (symptoms included pain at rest, pain during exertion, clicking, popping, locking, crepitus, and sensation of instability). MRI MRI was performed using a 1.5-Tesla MR system (MAG- NETOM, Symphony, Siemens Medical Systems, Erlangen, Germany) and a send/receive extremity radio-frequency coil. The knee was positioned in the extremity coil at an external rotation of approximately A routine MRI protocol of the knee was performed as follows: axial, T2-weighted spin echo (TR/TE, 3210/83 msec; 4-mm section thickness; 1.5-mm intersection gap; matrix size, ; field of view (FOV), ); sagittal, proton density-weighted (TR/TE, 2200/14 msec; 3-mm section thickness; 1.0-mm intersection gap; matrix size, ; FOV, ); sagittal, T2-weighted turbo spin echo (TR/effective TE, 3000/83 msec; 4-mm section thickness; 1.0-mm intersection gap; matrix size, ; FOV, ); coronal, proton density-weighted turbo spin echo with fat saturation (TR/effective TE, 2000/13 msec; 4-mm section thickness; 1.0-mm intersection gap; matrix size, ; FOV, ); and coronal, inversion recovery (TR/TE/TI, 4830/33/130 msec; 4-mm section thickness; 1.5-mm intersection gap; matrix size, ). The MRI examinations were conducted prior to competition and heavy training or at least 48 hours after to prevent exercise-induced changes from affecting interpretation of the MR findings. Interpretation of MR Examinations The MR examinations of the knees were reviewed separately, in a blinded manner by two radiologists, each with musculoskeletal MR experience, using established, previously described criteria (15 20). Attention was focused on identification and characterization of ligamentous abnormalities, meniscal abnormalities, articular cartilage lesions, hematopoietic hyperplasia, bone contusions, and other possible abnormalities that may affect the knee (15 20). This information was recorded on a standardized form to facilitate analysis. Additional assessment criteria used for specific findings were as follows: 1. Ligamentous abnormalities of the knee were characterized using standard interpretation criteria, as previously described (15 17), and were scored as normal, partial tear, or complete tear for the anterior cruciate ligament (ACL), posterior cruciate ligament (PCL), medial collateral ligament (MCL), and lateral collateral ligament (LCL). Additionally, if there was a ligament tear, it was designated as chronic or acute based on history and secondary findings. 2. The menisci were assessed for increased intrameniscal signal and morphologic alterations (15 17), using the grading scheme described by Mink et al (17): grade 0, normal; grade 1, intrameniscal globular focus of signal that is not adjacent to either the superior or inferior articular surface; grade 2, intrameniscal linear or wedge-shaped signal that does not extend to an articular surface; and grade 3, linear or globular signal communicating or extending to at least one articular surface. In general, grades 1 and 2 are indicative of meniscal degeneration, whereas grade 3 indicates a tear (15 17). 3. The articular cartilage was evaluated using an arthroscopic grading scheme that has been correlated with findings on MRI (15 17,21), as follows: grade 1 chondromalacia, focal areas of signal intensity changes cartilage surface or subchondral bone extension; grade 2 chondromalacia, blisterlike swelling with signal intensity changes extending to the articular cartilage surface; grade 3 chondromalacia, focal areas of signal intensity changes associated with a loss of sharp margins between articulating surfaces; and grade 4 chondromalacia, complete loss of articular cartilage, exposed subchondral bone, and underlying fluid in subchondral bone. 4. Hematopoietic bone marrow hyperplasia was characterized as regions of increased signal intensity on T2-weighted or proton density-weighted, fat-saturated pulse sequences or inversion recovery pulse sequences, or decreased signal intensity seen on T1-weighted MR images (22 24). The

3 124 Shellock et al. Table 1 Ironman Triathlete Study Subjects Demographic, Training, and Knee-Related Information* Total Group I Group II Study subjects Men Women Age 47 (35 66 years) 48 (37 66 years) 47 (35 62 years) No. of years training 12 (4 22) 11 (5 20) 13 (4 22) Total triathlons a 65 (7 175) 62 (14 150) 67 (7 175) Previous knee injury 16 (55%) 0 (0%) 16 (100%)* Site of injury Nonspecific b 6 (21%) 0 (0%) 6 (38%)* Lateral 2 (7%) 0 (0%) 2 (12%)* Medial 3 (10%) 0 (0%) 3 (19%)* Patellofemoral 5 (17%) 0 (0%) 5 (31%)* Current symptoms 11 (37%) 0 (0%) 11 (69%)* *Values indicated as mean and range. a Total number of triathlons competing in during lifetime. b Nonspecific refers to overall or unrelated to a specific site of the knee. epiphysis and proximal tibia were uninvolved with regard to the bone marrow alterations found associated with hematopoietic hyperplasia (24). Although hematopoietic hyperplasia is known to occur in both the proximal and distal femur (22 24), only changes on the distal portion of the femur were considered for the purpose of this study (24). 5. Bone bruises or contusions were identified based on the presence of poorly marginated signal intensity alterations (e.g., decreased signal intensity seen on T1-weighted pulse sequences, or increased signal intensity seen on T2-weighted or proton density-weighted, fat-saturated pulse sequences or inversion recovery pulse sequences) in the cancellous bone or marrow (15 19). 6. Effusions were qualitatively graded as small, moderate, or large (15 17). Data Analysis Data were analyzed using standard descriptive and statistical analysis techniques to evaluate the findings for the entire group of Ironman triathletes and to compare group I to group II. The appropriate statistical analysis was performed using StatView software (SAS Institute, Inc., Cary, NC). Statistically significant results were designated as those with a P value of 0.05 or less. RESULTS Table 1 summarizes the demographic, training, and knee-related information for the study subjects. For the entire group of triathletes (N 29), the average number of years of training was 12 years (range, 4 22 years) and the average number of triathlon competitions participated in per lifetime was 65 (range, 7 175) (Table 1). There were no statistically significant differences between the ages, number of years training, and number of triathlons participated in comparing the study subjects in group I (N 13) to group II (N 16). For group II (previous injury and/or current symptoms), 38% had an injury involving the overall aspects of the knee (i.e., nonspecific or unrelated to a specific site), 31% had an injury involving the patellofemoral joint, 19% had an injury involving the medial side of the knee, and 12% had an injury involving the lateral side of the knee. Additionally, 69% of the group II subjects were symptomatic during the period of their MRI examinations. Tables 2 and 3 show a summary of the MRI of the knee findings for the Ironman triathletes. For the entire group of study subjects, 10% (3/29) of the knees had ligamentous abnormalities (1, ACL, complete tear, chronic; 1, MCL, partial, chronic; 1, LCL, partial tear, chronic). The overall prevalence of abnormal ligaments was not statistically different comparing findings from group I (1/13, 8%) to group II (2/16, 12%) (P not significant). For the entire group of subjects, 55% (16/29) of the knees had abnormal menisci, as follows: grade 1, 7% (2/29); grade 2, 31% (9/29), and grade 3, 17% (5/29). There was a fair correlation between the age of the study subject and the presence of abnormal meniscal signal (regression analysis, R 0.4; P 0.05). The overall prevalence of abnormal menisci was significantly higher in group II (69%, 11/16) than in group I (38%, 5/13) (P 0.05). Cartilage abnormalities were found in 21% (6/29) of the entire group as follows: grade 1, 4% (4/29); grade 2, 3% (1/29); grade 3, 0% (0/29); and grade 4, 3% (1/29). The overall prevalence of chondromalacia was significantly higher in group II (31%, 5/16) than in group I (8%, 1/13) (P 0.05). Fourteen percent (4/29) of the knees of the study subjects had hematopoietic bone marrow hyperplasia (Fig. 1), with a prevalence that was not significantly different comparing group I (15%, 2/13) to group II (12%, 2/16) (P not significant). Twenty-one percent (6/29) of knees of the study subjects had bone contusions (Fig. 2). The overall prevalence of bone contusions was significantly higher in group II (31%, 5/16) than in group I (8%, 1/13) (P 0.05). Joint effusions were observed in 86% (25/29) of the knees of the study subjects (Fig. 3). The prevalence of joint effusions for group I (85%, 12/13) was comparable to that found in group II (81%, 13/16) (P not significant). The majority of the effusions were qualitatively

4 Knees of Ironman Triathletes: MRI Assessment 125 Table 2 MR Imaging of the Knee Findings for Ironman Triathlete Study Subjects Total Group I Group II Study subjects N 29 N 13 N 16 Ligaments ACL, normal 28 (97%) 13 (100%) 15 (94%) ACL, partial tear ACL, complete tear 1 (3%) 0 1 (6%) chronic PCL, normal 29 (100%) 13 (100%) 16 (100%) PCL, partial tear PCL, complete tear MCL, normal 28 (97%) 13 (100%) 15 (94%) MCL, partial tear 1 (3%) 0 1 (3%) chronic MCL, complete tear LCL, normal 28 (97%) 12 (92%) 0 LCL, partial tear 1 (3%) 1 (8%) chronic 0 LCL, complete tear Menisci Lateral, normal 27 (93%) 13 (100%) 14 (88%) Lateral, grade Lateral, grade 2 1 (3%) 0 1 (12%) Lateral, grade 3 1 (3%) 0 1 (12%) Medial, normal 15 (52%) 8 (61%) 7 (44%) a Medial, grade 1 2 (7%) 1 (8%) 1 (6%) Medial, grade 2 9 (31%) 3 (23%) 6 (37%) a Medial, grade 3 3 (10%) 1 (8%) 2 (13%) Cartilage Normal 23 (79%) 12 (92%) 11 (69%) a Grade 1 4 (14%) 0 4 (25%) a Grade 2 1 (3%) 1 (8%) 0 Grade Grade 4 1 (3%) 0 1 (6%) a P 0.05, Group I compared to group II. ACL anterior cruciate ligament; PCL posterior cruciate ligament; MCL medial collateral ligament; LCL lateral collateral ligament. categorized as small. Incidental findings included popliteal or Baker s cysts seen in 14% (3/29) of the knees of the study subjects. No stress fractures or other similar significant findings were observed. DISCUSSION Because of the prevalence of sports-related alterations in the knee joint, the correct and rapid differentiation between traumatic lesions, chronic degenerative changes, and the presence of possible adaptive processes is important with regard to the therapeutic management of athletes, as well as the guidance of proper training and competitive planning (10). Thus, the identification and characterization of conditions that affect the knees of active individuals is crucial. This may be optimally accomplished by evaluating the knee using MRI, which is known to have a high diagnostic reliabil- Table 3 MR Imaging of the Knee Findings for Ironman Triathlete Study Subjects Total Group I Group II Study subjects N 29 N 13 N 16 Hematopoietic hyperplasia No 25 (86%) 11 (85%) 14 (88%) Yes 4 (14%) 2 (15%) 2 (12%) Bone contusion No 23 (79%) 12 (92%) 11 (69%) a Yes 6 (21%) 1 (8%) 5 (31%) a Joint effusion No 4 (14%) 1 (8%) 3 (19%) Yes 25 (86%) 12 (85%) 13 (82%) 21 (84%) small 11 (92%) small 10 (77%) 2 (8%) moderate 0 (0%) moderate 2 (15%) 2 (8%) large 1 (8%) large 1 (8%) Cysts No 26 (86%) 12 (92%) 14 (88%) Yes 3 (14%) 1 (8%) 2 (12%) a P 0.05, group I compared to group II.

5 126 Shellock et al. Figure 1. MR image of the knee showing hematopoietic bone marrow hyperplasia (coronal, inversion recovery; TR/ TE/TI, 4830/33/130 msec). This MR image was obtained in a female Ironman triathlete with a history of injury involving the patellofemoral joint (15 years of training, competed in 120 triathlons). Note the area of increased signal intensity in the distal femur. ity for all essential structures (15 20). Additionally, it is useful to determine MRI appearances of the knee for asymptomatic and symptomatic athletic study populations to permit recognition of the spectrum of findings, and thus to avoid attributing a greater significance to these than is clinically justified (25). This is particularly important when observing findings that do not closely correlate with symptoms (25). The Ironman triathletes involved in this investigation were specifically selected for evaluation because they engaged in training and competition for at least 4 years and were older. Thus, they represent extreme examples of athletes that may have knee abnormalities related to excessive, repetitive musculoskeletal loads and stresses and/or the aging process. Interestingly, there were triathletes (group I) that reported no prior knee injuries or symptoms despite substantial participation in training (average of 11 years; range, 5 20 years) and competitive activities (average total number of triathlons, 62; range, ). Of note, the majority of triathletes with prior knee injuries (group II) reported experiencing either an overall knee problem (38%) or a patellofemoral joint condition (31%). With regard to findings for all ligaments of the knee, there was apparently a relatively low percentage (10%) of abnormalities for the older Ironman triathletes that were studied, with only one chronic, complete tear of the ACL (3/29) found (however, the incidence of chronic ligament abnormalities in the general population and athletic groups is unknown). The single complete ACL tear was identified in a 50-year-old triathlete with 20 years of training who had competed in 100 triathlons. Notably, there was no evidence of chondromalacia in this triathlete s knee, which is contrary to the belief that chronic incompetence of the ACL leads to progressive deterioration of knee function (3 6,9,15,17). With respect to the meniscal abnormalities found in the triathletes, in order to put this information into proper perspective, it is first necessary to understand the prevalence of meniscal abnormalities in asymptomatic and symptomatic populations of nonathletic and athletic individuals, especially with reference to agerelated changes (25 32). Meniscal tears are extremely common and may exist without associated signs and symptoms (26,30 32). Also, it is well known that degenerative meniscal changes are more common with increasing age, as shown by cadaveric and MRI-based studies (26,30,31). Kornick et al (26) used MRI of the knee to study asymptomatic volunteers (presumably active and inactive subjects were included) ranging in age from years old. There was at least a 25% prevalence of meniscal signal abnormalities seen as early as the second decade, which increased sharply with age. The prevalence of all signal abnormalities correlated well with age for this asymptomatic study population (26). By comparison, the overall findings for meniscal changes seen in the triathletes were comparable to those in the asymptomatic subjects of Kornick et al (26) in consideration of age-related alterations. With regard to athletic populations, Ludman et al (25) studied asymptomatic gymnasts aged years old. There was a 13% incidence of grade 3 changes for this study group (25). In an MRI study of collegiate football and professional basketball players (25% had previous surgery or knee injuries), Brunner et al (27) reported that 55% of the asymptomatic subjects had substantial (grades 2 and 3) signal intensity abnormalities. Jerosch

6 Knees of Ironman Triathletes: MRI Assessment 127 Figure 2. MR image of the knee (coronal, inversion recovery; TR/TE/TI, 4830/ 33/130 msec) showing bone contusion involving the tibial plateau of an asymptomatic male Ironman triathlete with no history of injury (16 years of training, competed in 100 triathlons). et al (28) used MRI to study the menisci of asymptomatic athletic subjects ranging in age from 8 62 years. There was an increase in meniscal degeneration associated with age, with substantial meniscal lesions (i.e., tears) seen in a significant number of asymptomatic subjects, especially those older than 50 years (28). In a study of high-endurance subjects similar to the triathletes of the present study, Shellock et al (29) used MRI of the knee to evaluate asymptomatic marathon runners without prior injuries or surgery (average age, 40 years old; average number of years training, 10 years). The overall prevalence of meniscal tears was 9%, with a 6% prevalence in subjects less than 45 years old and a 14% prevalence in marathon runners 45 years old or older (29). Notably, there was no relationship between the age of the runner and the presence of abnormal meniscal signal (29). In the present study, 55% of the knees had increased signal intensity changes (grades 1 3), with a significantly (P 0.05) higher prevalence found in group II subjects than in group subjects I. Additionally, the presence of abnormal meniscal signal correlated with the age of the triathlete. Substantial (i.e., grade 3) changes occurred in 14% of all triathletes, distributed as 8% in subjects that were asymptomatic (group I) and 19% in subjects that were symptomatic or had previous injuries (group II). Thus, in consideration of the various reports for age-related meniscal alterations found in asymptomatic and symptomatic, nonathletic and athletic populations, the older, competitive triathletes did not have substantially higher prevalence of abnormalities. This baseline presence of abnormal meniscal signal changes is especially crucial to recognize in order to prevent overreading of MRI of the knee for this active population (25 29). In general, there is controversy regarding the effects of overuse activities or repetitive impulse loading on soft tissues and osseous structures (1 9,25 42). For example, opinions vary regarding the effects of running on meniscal damage and the development of osteoarthritis (29,32 42). Interestingly, the findings for the menisci of the triathletes suggest that despite being older and having participated in high-level training and competition, these structures may have undergone compensatory adaptations, similar to what has been suggested to occur for marathon runners (29). Cartilage abnormalities were found in 21% (6/29) of all triathletes, with a higher prevalence in group II than in group I. This was not surprising considering that history of an injury is associated with the development

7 128 Shellock et al. Figure 3. MR image of the knee (axial, T2-weighted spin echo; TR/TE, 3210/83 msec) obtained in a male Ironman triathlete with a previous overuse injury (16 years of training, competed in 100 triathlons). Note the presence of a moderate-sized effusion (the contiguous section locations also showed joint fluid). of osteoarthritis (1 9). Of note is that the majority of the cartilage lesions found in the triathletes were relatively minor (83%, 5/6, grades 1 and 2). The prevalence of chondromalacia affecting the knee based on an MRI assessment is unknown. Therefore, the importance of these data has not been determined. Hematopoietic bone marrow hyperplasia has been detected incidentally in the distal femurs of patients during routine MRI of the knee (22). This finding was considered to be a benign process observed mostly in mildly to moderately obese women, some of whom were smokers with associated peripheral leukocytosis (relationship unknown) (22). A study conducted by Shellock et al (24) reported a prevalence of hematopoietic hyperplasia of 3% in healthy volunteers and 43% in asymptomatic marathon runners (P 0.05). It was postulated that the high prevalence of hematopoietic bone marrow hyperplasia found in marathon runners was a response to sports anemia (24), which is commonly found in highly conditioned, aerobically trained athletes (43 46), and thus a normal variant finding. Therefore, the conversion of fatty bone marrow to hematopoietic marrow may be a reaction to sports anemia, similar to what is observed with other forms of anemia (45). Caldemeyer et al (46) studied the frequency of hematopoietic hyperplasia on spinal MR images obtained in endurance athletes and correlated this information with clinical parameters. Fifty-three percent (8/15) of the study subjects showed evidence of hematopoietic hyperplasia (46). Notably, there was no correlation between this finding and the duration of training, hematologic results, or maximal oxygen consumption levels. There were borderline relationships between hematopoietic hyperplasia and the presence of anemia and the intensity of training (46). The investigators concluded that decreased iron reserves or increased hematopoiesis probably contributed to hematopoietic hyperplasia seen in this population of endurance athletes (46). In the present study, hematopoietic bone marrow hyperplasia was found in 14% of all triathletes, without a significant difference in prevalence of this finding between group I (15%) and group II (12%) subjects. Interestingly, this result is higher than that reported for healthy volunteer subjects, but less than that reported for other endurance-trained athletes (24,46). The significance of this requires additional investigation, which will likely entail measurement of physiologic and hematologic parameters correlated with MRI findings of the knee. Bone marrow contusions are frequently identified at MRI after an injury to the musculoskeletal system (15 20). These osseous abnormalities may result from a direct blow to the bone, from compressive forces of adjacent bones impacting one another, or from traction forces that occur during an avulsion injury (15 20). Additionally, there are other causes of bone contusions unrelated to direct trauma (17). In this study, 21% (6/29) of the knees of the Ironman triathletes had bone contusions, with a higher prevalence (31%) occurring in the group II triathletes (P 0.05). This greater prevalence is understandable in consideration of the fact that all group II triathletes had a history of a previous injury and/or symptoms during the MRI examination of the knee. For the five patients in group II with bone contusions, four suffered prior trauma and one had painful symptoms at rest and

8 Knees of Ironman Triathletes: MRI Assessment 129 during activity. Unfortunately, we were unable to attribute an exact etiology to the bone contusion findings in these subjects. With regard to joint effusions, normal knee contains 4 ml or less of synovial fluid, and MRI is capable of detecting as little as 1 ml of fluid within this joint (15). Hill et al (47) used MRI of the knee to study the prevalence of effusions and other abnormal findings in older subjects with and without knee symptoms. Moderate to large effusions were seen in 11% of the subjects without symptoms and in 55% of the subjects with knee pain. Furthermore, Hill et al (47) reported that effusions tended to be more common in middle-aged and elderly people, with moderate or large effusions observed in those with knee pain compared to those without. No similar MRI investigation has been performed in an athletic population, so the relative incidence of knee effusions in physically active individuals is unknown. In the triathletes of this study, knee joint effusions were observed in a high percentage (86%) of the overall study group, but the majority (84%) of these were qualitatively categorized as small or, importantly, within the aforementioned normal limits for knee joint fluid. There was no statistically significant difference between the presence of fluid for group I vs. group II triathletes. Moderate or large effusions were seen in 19% (3/16) of subjects with prior injuries and/or current symptoms, which is the expected prevalence for a population of older, symptomatic subjects (36). Popliteal or Baker s cysts are typically found in adult knees in association with internal derangements, osteoarthritis, rheumatoid arthritis, chondromalacia, granulomatous synovitis, osteochondritis dissecans, and other abnormal conditions (15,17). Cysts were seen in 14% of the triathlete knees. Two of these subjects (both in group II) had internal derangements (one, partial tear, MCL; one, grade 2 abnormal meniscal signal), and the other had an entirely normal knee. The importance of these incidental findings in the older triathlete knee is unknown. In summary, a spectrum of findings on MRI of the knee for older, competitive Ironman triathletes without (group I) and with (group II) prior injuries and/or symptoms has been described. There was a higher prevalence of meniscal abnormalities, chondromalacia, and bone contusions observed for triathletes in group II. In general, the presence of substantial findings was no greater than that reported for age-related changes previously reported for other nonathletes and athletes. These results have important implications for the diagnostic use of MRI of knee in this particular athletic population. REFERENCES 1. Sutton AJ, Muir KR, Mockett S, Fentem P. A case-control study to investigate the relation between low and moderate levels of physical activity and osteoarthritis of the knee using data collected as part of the Allied Dunbar National Fitness Survey. Ann Rheum Dis 2001; 60: Buckwalter JA, Lane NE. Athletics and osteoarthritis. Am J Sports Med 1997;25: Sohn RS, Micheli LJ. The effect of running on the pathogenesis of osteoarthritis of the hips and knees. Clin Orthop 1985;198: Ward MM, Hubert HB, Shi H, Block DA. Physical disability in older runners: prevalence, risk factors, and progression with age. J Gerontol A Biol Sci Med Sci 1995;50:M70 M Saxon L, Finch C, Bass S. Sports participation, sports injuries and osteoarthritis: implications for prevention. Sports Med 1999;28: Vad VB, Bhat AL. The athlete with early knee arthritis. Phys Med Rehabil Clin N Am 2000;11: Kuipers H, Keizer HA, Schep G, Hoogeveen AR. Long-term intensive athletic training: few adverse effects on later physical health. Ned Tijdschr Geneeskd 2001;145: Neidhard M, Muller-Ladner U, Frey W, et al. Increased serum levels of non-collagenous matrix proteins (cartilage oligomeric matrix protein and melanoma inhibitory activity) in marathon runners. Osteoarthritis Cartilage 2000;8: Spector TD, Harris PA, Hart DJ, et al. Risk of osteoarthritis associated with long-term weight-bearing sports: a radiologic survey of the hips and knees in female ex-athletes and population controls. Arthritis Rheum 1996;39: Wacker F, Bloze X, Mellerowicz H, Wolf KJ. Diagnosis of changes in the knee joint of high performance athletes. Radiologe 1995;35: Collins K, Wagner M, Peterson K, Storey M. Overuse injuries in triathletes. A study of the 1986 Seafair Triathlon. Am J Sports Med 1989;17: Clements K, Yates B, Curran M. The prevalence of chronic knee injury in triathletes. Br J Sports Med 1999;33: Vleck VE, Garbutt G. Injury and training characteristics of male elite, development squad, and club triathletes. Int J Sports Med 1998;19: Korkia PK, Tunstall-Pedoe DS, Maffulli N. An epidemiological investigation of training and injury patterns in British triathletes. Br J Sports Med 1994;28: Stoller DA, Cannon WD, Anderson LJ. The knee. In: Stoller DA, editor. Magnetic resonance imaging in orthopaedics and sports medicine, 2nd edition. Philadelphia: Lippincott-Raven Publishers; p Crues JV, Shellock FG. MRI of the knee. In: Stark DD, Bradley WG, editors. Magnetic resonance imaging: a comprehensive text, 3rd edition. New York: Mosby Co.; p Mink JH, Reicher MA, Crues JV, Deutsch AL. MRI of the knee, 2nd edition, New York: Raven Press; Deutsch AL, Mink JH, Shellock FG. Magnetic resonance imaging of injuries to bone and articular cartilage: emphasis on radiographically occult abnormalities. Orthoped Rev 1990;9: Sanders TG, Medynski MA, Feller JF, Lawhorn KW. Bone contusion patterns of the knee at MR imaging: footprint of the mechanism of injury. Radiographics 2000;20:S135 S Lynch TC, Crues III JV, Morgan FW, et al. Bone abnormalities of the knee: prevalence and significance at MR imaging. Radiology 1989; 171: Conway WF, Hayes CW, Loughran T, et al. Cross-sectional imaging of the patellofemoral joint and surrounding structures. Radiographics 1991;11: Deutsch AL, Mink JH, Rosenfeld RP, Waxman AD. Incidental detection of hematopoietic hyperplasia on routine knee MR imaging. AJR Am J Roentgenol 1989;152: Dooms GC, Fisher MR, Hricak F, Richardson M, Crooks LE, Genant HH. Bone marrow imaging: magnetic resonance studies related to age and sex. Radiology 1985;155: Shellock FG, Morris E, Deutsch AL, Mink JH, Kerr R, Boden SD. Hematopoietic bone marrow hyperplasia: high prevalence on MR images of the knee in asymptomatic marathon runners. AJR Am J Roentgenol 1992;158: Ludman CN, Hough DO, Cooper TG, Gottschalk A. Silent meniscal abnormalities in athletes: magnetic resonance imaging of asymptomatic competitive gymnasts. Br J Sports Med 1999;33: Kornick J, Trefelner E, McCarthy S, Lange R, Lynch K, Jokl P. Meniscal abnormalities in the asymptomatic population at MR imaging. Radiology 1990;177: Brunner MC, Flower SP, Evanch AM, Allman FL, Apple DF, Fajman WA. MRI of the athletic knee: findings in asymptomatic professional basketball and collegiate football players. Invest Radiol 1989;24: Jerosch J, Hoffstetter I, Reer R, Assheuer J. Strain-related longterm changes in the menisci in asymptomatic athletes. Knee Surg Sports Traumatol Arthrosc 1994;2:8 13.

9 130 Shellock et al. 29. Shellock FG, Deutsch A, Mink JH, Kerr R. Do asymptomatic marathon runners have an increased prevalence of meniscal abnormalities? An MR study of the knee in 23 volunteers. AJR Am J Roentgenol 1991;157: Noble J, Hamblen DL. The pathology of the degenerate meniscus lesion. J Bone Joint Surg 1975;57B: Noble J. Lesions of the menisci. Autopsy incidence in adults less than fifty-five years old. J Bone Joint Surg 1977;59A: Casscells SW. A postmortem study of the degenerate meniscus lesion. J Bone Joint Surg 1969;51A: Shellock FG, Mink JH. Knees of trained long-distance runners: MR imaging before and after competition. Radiology 1991;79: Kursunoglu-Brahme S, Schwaighofer B, Gundry C, Ho C, Resnick D. Jogging causes acute changes in the knee joint: an MR study of normal volunteers. AJR Am J Roentgenol 1990;154: Merkel KH. The surface of human menisci and its alterations during age. Arch Orthop Trauma Surg 1980;97: Radin EL, Paul IL, Rose RM. Role of mechanical factors in pathogenesis of primary arthritis. Lancet 1972;1: Sohn RS, Micheli LJ. The effect of running on the pathogenesis of osteoarthritis of the hips and knees. Clin Orthop 1985;198: Lane NE, Bloch DA, Jones HH, Marshall WH, Wood PD, Fries JF. Long-distance running, bone density, and osteoarthritis. JAMA 1986;255: Panush RS, Schmidt C, Caldwell JR, Edwards NL, Longley S, Yonker R, et al. Is running associated with degenerative joint disease? JAMA 1986;255: McDermott M, Freyne P. Osteoarthritis in runners with knee pain. Br J Sports Med 1983;17: Konradsen L, Hansen E-MB, Sondergaard L. Long distance running and osteoarthritis. Am J Sports Med 1990;18: Reimann I. Experimental osteoarthritis of the knee in rabbits induced by alteration of the loading bearing. Acta Ortho Scand 1973; 44: Carlson DL, Mawdsley RH. Sports anemia: a review of the literature. Am J Sports Med 1986;14: Balaban EP, Cox JV, Snell P, Vaughn RH, Frenkel EP. The frequency of anemia and iron deficiency in the runner. Med Sci Sports Exer 1989;21: Steiner RM, Mitchell DG, Rao VM, et al. Magnetic resonance imaging of bone marrow: diagnostic value in diffuse hematologic disorders. Magn Reson Q 1990;6: Caldemeyer KS, Smith RR, Harris A, et al. Hematopoietic bone marrow hyperplasia: correlation of spinal MR findings, hematologic parameters, and bone mineral density in endurance athletes. Radiology 1996;198: Hill CL, Gale DG, Chaisson CE, et al. Knee effusions, popliteal cysts, and synovial thickening: association with knee pain in osteoarthritis. J Rheumatol 2001;28: