Osgood-Schlatter disease was named after two physicians, Robert Osgood and Carl Schlatter, who, in 1903, independently described a partial avulsion of the tibial tubercle occurring during adolescence1,2. In boys, the condition appears between the ages of ten and fifteen years and especially affects those actively participating in sports3. Ogden and Southwick proposed that Osgood-Schlatter disease is caused by avulsion of the secondary ossification center of the tibial tuberosity4, and this theory is generally accepted. Other possible causes, such as an insult to the patellar tendon and associated soft tissues or an apophyseal lesion related to patella alta, have also been considered5,6. Recent research has shown that patients with Osgood-Schlatter disease exhibit elongated patellae and patellar tendons, which may result from long-standing tension on the extensor apparatus during a growth spurt, when femoral growth exceeds that of the anterior structures of the knee7. Osgood-Schlatter disease is characterized by pain, swelling, and tenderness in the anterior aspect of the proximal part of the tibia.
Previous studies, although quite limited in number, have described the relationship of Osgood-Schlatter disease to, and its effects on, sports activity by adolescents8; surgical procedures used for Osgood-Schlatter disease9-11; or complications of the disease12-14. Nonsurgical methods have been the most widely used primary form of treatment15. It has been claimed that, in the majority of adolescents, symptoms completely resolve after conservative treatment consisting of activity modification, application of ice, use of anti-inflammatory agents, and time16. However, in a study of nonsurgical treatment of skeletally immature patients, 60% still experienced discomfort with kneeling at the time of a nine-year follow-up17.
Surgical treatment of Osgood-Schlatter disease, albeit rare, may occasionally be warranted if disabling symptoms persist, but to our knowledge no previous study has addressed the long-term outcome of surgery in skeletally mature young adults. Binazzi et al. examined the results of twenty-six knees, in both skeletally immature and mature patients, at an average of thirteen years after surgical treatment18. They found that excision of the ossicles, with or without removal of the prominent tibial tubercle, clearly yielded better results than did various other procedures. Flowers and Bhadreshwar reported relief of pain in 95% of forty-two knees, in both skeletally immature and mature patients, at an average of five years after surgery9. Orava et al., who also included both skeletally immature and skeletally mature patients in their study, found that the outcomes of 80% of seventy operations were good after a duration of follow-up of 2.2 years19. Knowing the long-term results of surgical treatment of symptomatic Osgood-Schlatter disease in skeletally mature patients would be helpful to orthopaedic surgeons treating this population.
The purpose of the present study was to assess the rate of occurrence of surgical treatment of unresolved Osgood-Schlatter disease as well as the clinical course, radiographic characteristics, and long-term outcome following that treatment in a large population of young male military recruits.
Eligible patients who had been treated surgically for unresolved Osgood-Schlatter disease were identified from the medical records of the main military hospital where all surgical procedures on recruits performing compulsory military service in Finland were carried out between 1983 and 1995. A computer search for procedures done from 1983 through 1995 was performed with use of the appropriate diagnostic codes of the Eighth (1969 through 1986) and the Ninth (1987 through 1995) Revision of the International Classification of Diseases (ICD). We identified 178 consecutive young male recruits who had undergone surgery for unresolved Osgood-Schlatter disease. The original, complete medical records, including radiographs, were retrieved and reviewed to confirm the accuracy of the diagnoses and to systematically collect data for the present study.
In Finland, military service is compulsory for all male citizens. During the study period, from 1983 through 1995, the number of recruits starting their service annually ranged from 24,739 to 38,823 (mean, 32,258), and >80% of men between the ages of eighteen and twenty-nine years (median, nineteen years) completed a service period ranging from eight to eleven months. The study population consisted of 423,120 military recruits at risk nationwide. Combat training, marching, and other physical training activities were performed almost daily throughout the service period. A prominent tibial tubercle and an anterior ossicle are disorders that may interfere with the performance of military training because kneeling and crawling cause pain.
Primarily, the patients with symptomatic Osgood-Schlatter disease were treated conservatively with anti-inflammatory medications in their units and temporarily exempted from all strenuous physical activity and military training. No outpatient data concerning the prevalence of Osgood-Schlatter disease between 1983 and 1995 are available from garrisons, but since it has been estimated that one-fifth of adolescents actively participating in sports have symptoms of Osgood-Schlatter disease8, it can be estimated that thousands of individuals in the current study population of military trainees had symptoms of Osgood-Schlatter disease. Patients in whom the Osgood-Schlatter disease did not respond to conservative treatment and was restricting their training were referred by the physicians in the basic military units for orthopaedic consultation at the main military hospital to assess the need for surgery. The decision to operate was based on mutual agreement between the patient and the surgeon. Surgical treatment was suggested if (1) the patient had radiographic and clinical evidence of Osgood-Schlatter disease; (2) the duration of symptoms (median, four years [range, one to fourteen years] in the present study), including that before the compulsory military service period, was considered to be long enough to demonstrate the severity of the disease (Table I); (3) the patient found it impossible to continue military training because of the ineffectiveness of conservative treatment; and (4) the patient was unable to kneel or squat without persistent pain during the military service. The operations were performed at Central Military Hospital, which provided all surgical services for the entire armed forces of the country at that time.
Surgical Technique and Postoperative Treatment
Two alternative surgical approaches were used during the study period. In eighty-seven knees, a vertical 5-cm incision was made over the center of the distal part of the patellar tendon at a site 1 cm proximal to the tibial tubercle. The distal part of the patellar tendon was divided longitudinally. Any prominent tibial tubercle bone was removed with use of an osteotomy, and posterior intratendinous ossicles were excised when present (in 109 knees in the series as a whole). With use of this approach, the most common operative technique was to remove the ossicle(s) with or without resecting the uneven surface of the tibial tubercle. With the alternative method (used in thirty knees), a transverse 5-cm incision was centered over a point 1 cm proximal to the tibial tubercle. The lateral soft-tissue attachments of the patellar tendon were released longitudinally, leaving the patellar tendon intact. The tendon was elevated, and intratendinous, osteocartilaginous fragments were removed. The tibial tubercle was resected down to the insertion of the tendon and smoothed with a file. Tibial tubercle resection and ossicle removal were done in eighty-eight knees, ossicle removal without tibial tubercle resection was performed in twenty-one knees, and tibial tubercle resection only was carried out in eight knees. No drilling was performed with either method. A light compression dressing was applied to the whole lower extremity. Quadriceps-setting exercises were started on the first postoperative day. The postoperative mobilization was supervised by a physiotherapist. After regaining adequate quadriceps function, the patient returned to his unit, and he was exempted from all strenuous physical activity and military training for six to twelve weeks. Crutches were prescribed for a short postoperative period if needed. All patients in this sample completed their full-time military service.
Collection of Follow-up Data
All 178 military recruits who were treated surgically for Osgood-Schlatter disease between 1983 and 1995 were subsequently invited to return for physical and radiographic examinations at the outpatient department of the main military hospital. If the patient did not return for follow-up after the first or second invitation letter, his mailing address was checked and a third, final invitation letter was mailed.
At the final follow-up examination, the participants were asked whether they had received treatment for Osgood-Schlatter disease or some other disorder of the same knee in another hospital after the compulsory military service period. When the participant responded affirmatively, the appropriate medical records and radiographs from the hospitals where the treatment had been carried out were retrieved for review and analysis. The inclusion criterion for the present study was surgically treated Osgood-Schlatter disease during the period of compulsory military service. The exclusion criteria were surgery on the same knee after the index operation for reasons other than Osgood-Schlatter disease, a known disease such as rheumatoid arthritis or infection, or trauma involving the knee. On the basis of these criteria, thirteen patients who reported having undergone surgery on the same knee after the index surgery for reasons other than Osgood-Schlatter disease were excluded from the long-term outcome analysis. Of the 165 remaining possible participants, 107 (65%) agreed to participate in the follow-up study. The reasons for nonparticipation were unknown. Data collected from individuals participating in the final follow-up examination were used for the long-term analysis.
The final follow-up examination consisted of administration of a questionnaire and interviews to determine the patient's functional outcome after surgery (according to the modified Kujala20 score and a visual analog pain score, whether the patient regularly participated in sports, how he coped with daily activities, and his kneeling and squatting ability) and a physical examination by an orthopaedic surgeon complemented by anteroposterior, lateral, and axial radiographs of the surgically treated knee in 30° of flexion (the Laurin projection21). All radiographs for the 107 patients (117 knees) were available for review.
The status of the operatively treated knee was evaluated by physical examination and use of the modified Kujala score20. The question concerning patellar dislocations in the original Kujala score was replaced with a question assessing difficulty with kneeling, with five possible responses: no problems, repeated or prolonged kneeling causes pain, kneeling is painful every time, can kneel with help, and impossible to kneel. Thus, the total summed score, 100 points, remained the same as that in the original Kujala score. The results of the surgery were classified as excellent (a Kujala score of 95 to 100 points), good (85 to 94 points), fair (65 to 84 points), or poor (less than 65 points) at the follow-up examination20,22. The Kujala score has been validated for evaluation of anterior knee pain and is often used to assess patellofemoral problems, but it has not been validated specifically for problems related to the tibial tubercle8,20. Because of the lack of a specific score for problems related to the tibial tubercle, the symptom of Osgood-Schlatter disease can also be considered to be anterior knee pain. A 0 to 100-mm visual analog scale, with 0 denoting no pain, 1 to 30 mm indicating mild pain, 31 to 60 mm indicating moderate pain, 61 to 99 mm indicating severe pain, and 100 mm denoting the worst imaginable pain, was used to assess the degree of subjective pain experienced by the patient during normal activities during the week before the last follow-up examination. The intensity of pain at the time of follow-up was considered to be low when the visual analog pain score did not exceed 30 mm.
Radiographs made preoperatively and at the time of final follow-up were used to measure the proximal position of the patella with use of two patellar indexes: the Insall-Salvati method23 (the ratio of the length of the patella to the length of the patellar tendon) (Fig. 1, A) and the Blackburne-Peel method24 (the ratio of the perpendicular height of the distal articular surface of the patella from the line of the tibial plateau to the length of the patellar articular surface) (Fig. 1, B). To assess changes in the area of the tibial tuberosity after surgery, the tibial tubercle prominence (referred to here as the tibial tuberosity index) was measured as the ratio of the thickness of the tuberosity to the distance between the top of the tuberosity and the middle vertical line of the tibia. The line through the base of the tibial tuberosity is parallel to the mid-vertical tibial line. The mid-vertical tibial line was determined by measuring the middle of the projection of the tibia from four points located at various perpendicular levels of the cortex of the proximal part of the tibial shaft. This line was used to avoid any measurement error caused by the various curvatures of the posterior tibial cortex. The height of the tuberosity was measured from a line running parallel to the mid-vertical tibial line and passing through the base of the tuberosity. The base of the tubercle was determined by adjusting the line through the estimated base of the tibial tuberosity so that it was parallel to the mid-vertical tibial line and delineated the tibial tuberosity from the anterior tibial cortex7. The thickness of the tuberosity was measured from the top of the tuberosity to the line through the base of the tuberosity (the line parallel to the mid-vertical tibial line)7 (Fig. 1, C). The tibial tuberosity index was measured on the preoperative radiographs and those made at the time of final follow-up (Fig. 2) to assess the effect of surgery on the thickness of the osseous prominence. All radiographic results were interpreted by a musculoskeletal radiologist.
Statistical Analysis
The series was analyzed according to the number of patients with one operatively treated knee (ninety-seven patients) or two operatively treated knees (ten patients), depending on the outcome variable. The association between the outcome of the surgery and the surgical approach, duration of symptoms, and age at the onset of symptoms was assessed. The t test was used when we analyzed continuous, unskewed data, whereas the Mann-Whitney U test was used for continuous, skewed data. The t test was used to analyze the association between the number of operatively treated knees per person and the visual analog pain score. Correlation coefficients were calculated when we explored the associations between continuous variables (radiographic indexes and the Kujala score). The difference between two independent groups was considered significant if p = 0.05.
Source of Funding
There was no external funding source for the investigation.
Occurrence
Of all of the male recruits performing their compulsory military service during the thirteen-year period from 1983 to 1995, 178 underwent surgery for unresolved Osgood-Schlatter disease. The median age of the patients at the time of surgery was twenty years (range, eighteen to twenty-nine years). The person-based occurrence of surgically treated Osgood-Schlatter disease was forty-two per 100,000 recruits. Our follow-up sample included 107 patients treated with a total of 117 operations (ninety-seven unilateral and ten bilateral). The median age at the onset of symptoms was fifteen years (range, seven to twenty-three years). The median time from the onset of symptoms to the operation was four years (Table I).
Postoperative Complications and Outcomes at the Time of Final Follow-up
Postoperative complications were detected in six patients, and they included three infections, two hematomas, and one deep venous thrombosis. The median follow-up time after the surgery was ten years (range, six to nineteen years). The median modified Kujala score for the patients who returned for the follow-up visit was 95 points (range, 35 to 100 points). The outcome after surgical treatment was excellent (a score of >94 points) for fifty-five patients (51%), good for thirty-six (34%), and poor for sixteen (15%) (Table II). The median visual analog score indicating the intensity of pain during the week prior to the time of follow-up was 7 mm (range, 0 to 82 mm). Thirty-seven knees (35%) were completely painless, the score was =30 mm for fifty-three knees (50%), and the score was >60 mm for six knees (6%).
Forty-one patients (38%) reported no problems when kneeling, and twenty-five patients (23%) reported pain associated with repeated or prolonged kneeling (Table II). Seventy-three patients (68%) reported that squatting did not cause problems. Ninety-three patients (87%) had no restrictions in everyday activities or at work, six patients (6%) had some restrictions at work, and seven patients (7%) were compelled to change employment because of knee symptoms. One patient, with a modified Kujala score of 94 points, had retired because of other health problems. Eighty patients (75%) had resumed the same level of leisure-time sports activity with the same degree of participation as before the operation, sixteen (15%) had reduced their sports participation because of the operation, and three (3%) were unable to participate in any sports activity after the surgery.
With the numbers studied, the surgical approach (a vertical or transverse incision) was not found to be associated with the clinical outcome of the surgery as measured with the modified Kujala score (p = 0.45). Furthermore, we could not identify an association between the duration of symptoms (p = 0.45) or the age at the onset of symptoms (p = 0.38) and the outcome of the surgery. Patients who had undergone a bilateral operation for treatment of Osgood-Schlatter disease reported a higher median visual analog pain score than did patients with one operatively treated knee (29 and 5 mm, respectively; p = 0.09). Only two patients had subsequently undergone a reoperation for treatment of the Osgood-Schlatter disease.
Radiographic Findings
Preoperatively, ossicles were observed radiographically in 109 knees (93%). All ossicles were removed during the operation, but new ossicles were seen in forty-four knees (38%) at the time of final follow-up. The presence of ossicles at the time of follow-up was not related to the outcome of the surgery as measured with the modified Kujala score (p = 0.96) or the visual analog pain scale (p = 0.54). The average tibial tuberosity index was found to have decreased by 47% (range, 17% to 65%) when the preoperative and postoperative radiographs were compared. However, we could not identify an association between the postoperative tibial tuberosity index and the outcome of the surgery (p = 0.75). Preoperative assessments according to the Insall-Salvati method showed the patellar heights to be close to normal (mean Insall-Salvati index, 1.0), and patella alta (an Insall-Salvati index of >1.2) was found in only three (3%) of the knees. There were also three knees with a low-riding patella (an Insall-Salvati index of <0.8). Postoperatively, the mean Insall-Salvati index was 1.09, indicating a relative lengthening of the patellar tendon of 10% (p = 0.003). Twenty-four knees (21%) had patella alta, and only one knee (1%) had a low-riding patella. The mean Blackburne-Peel index was 0.85 preoperatively and 0.95 postoperatively (p = 0.003). The preoperative Blackburne-Peel indexes were close to normal, with ten knees (9%) showing patella alta (>1.0) and only one knee (1%) showing patella baja. Neither the preoperative nor the postoperative Insall-Salvati index, nor the change in the index after the surgery, was related to the modified Kujala score (p = 0.21). Also, no relationship was found between the tibial tuberosity index (p = 0.75) or the Blackburne-Peel index (p = 0.99) and the surgical outcome as measured with the modified Kujala score.
The principal finding of this study was that surgical treatment of unresolved Osgood-Schlatter disease in skeletally mature young adults does not appear to have long-term deleterious effects. Secondly, unresolved Osgood-Schlatter disease with persistent symptoms indicating a need for surgical treatment was a rare condition in a large population-based sample of young men. The long-term outcome of surgical treatment of Osgood-Schlatter disease was good with respect to the modified Kujala score, maintenance of physical activity level, and residual pain intensity. After surgical treatment, a majority of the patients reported no restrictions in everyday activities either at home or at work. More than three-fourths of the patients were able to resume the same level of sports activity and degree of participation as before the operation, and the level was maintained over the ten-year follow-up period. However, even after surgery, one of the main symptoms of Osgood-Schlatter disease—i.e., pain when kneeling—was reported by 39% of the patients, although the Kujala score and the visual analog pain score were good. This rate of pain when kneeling was markedly lower than that in the report by Krause et al.17, in which 60% of the patients experienced discomfort with kneeling at nine years following nonsurgical treatment for Osgood-Schlatter disease. However, comparisons with that study should be made with caution because the patients differed substantially, at least in terms of their age and the variety of their symptoms, from our patients. In conclusion, it seems that neither conservative nor surgical treatment successfully eliminates one of the main symptoms of Osgood-Schlatter disease (pain with kneeling) in all patients.
To our knowledge, our series, in which the median duration of follow-up was ten years, is the largest retrospectively collected homogeneous sample of young, healthy, skeletally mature men. Because of differences in study settings and samples, caution must be exercised when comparing our findings with those in other studies of the results of operative treatment of Osgood-Schlatter disease. The present series consisted of severe, symptomatic, unresolved cases of Osgood-Schlatter disease for which nonoperative treatment had failed and symptoms had worsened as a result of strenuous military training. The median duration of symptoms in our patients was four years before the operation. When these facts are taken into account, the median modified Kujala score for our patients at the time of follow-up (95 points) can be considered to be excellent. This score corresponds well to the results in the study by Orava et al., who reported excellent or good results in physically active young adult patients, although the average follow-up period was only 2.2 years in that study and some of their patients were skeletally immature19. Similarly, good clinical scores were reported by Flowers and Bhadreshwar9, in whose study 95% of forty-two knees had relief of pain at the time of follow-up, at an average of five years after surgery9. However, all of their patients were skeletally immature and thus not directly comparable with our patients.
Our patients' physical activity levels at the time of follow-up can be considered good; 87% had no restrictions in everyday life, and 80% had resumed the same level of leisure-time sports activity and participation as before the operation. In fact, the physical activity levels of our patients correspond to the general activity levels of thirty-one-year-old adults in Finland, 60% of whom reportedly participate in sports at least once a week25. The activity levels of our patients are also in agreement with those seen after nonsurgical treatment in the study by Krause et al., who stated that 76% of patients between the ages of sixteen and thirty-three years at the end of the follow-up period reported no limitation in daily activities17.
The rarity of unresolved Osgood-Schlatter disease was confirmed in the current study. Although the study population consisted of approximately 400,000 military recruits nationwide, with new batches entering the mandatory military training program annually, it was necessary to review the cases seen over thirteen years to accumulate an appropriate number of patients with unresolved Osgood-Schlatter disease for a long-term follow-up study. In the present series, the occurrence of operatively treated, unresolved Osgood-Schlatter disease was forty-two per 100,000 recruits per year. The proportions of knees treated surgically for Osgood-Schlatter disease in previous, larger series, with twenty to seventy patients, have ranged from approximately 10% to 12% of the total number of patients with the disease3,26, but no population-based occurrence values have previously been published, to our knowledge. In a study of 389 adolescents, Kujala et al. reported Osgood-Schlatter disease in 21% of those who actively participated in sports as compared with only 4.5% of those who did not8. On the basis of these percentages, the overall population at risk for surgical treatment of Osgood-Schlatter disease in the current study included at least thousands of patients with the disease. To put these occurrence rates in some perspective, the reported occurrence of operatively treated bipartite patella is nine per 100,000 recruits per year22, the reported incidence of acute traumatic patellar dislocations is seventy-seven per 100,000 recruits per year27, and the reported occurrence of displaced fatigue fractures of the femoral neck is three per 100,000 recruits per year28.
In the present study, ossicles were observed radiographically in a large majority (93%) of the knees, indicating the severity of the disease in our sample. It was interesting that, although we believe that all ossicles were removed during the index surgery, one-third of the knees in the present study were seen to have ossicles on plain radiographs made at the time of follow-up. In these cases, ossification may have occurred within the tendon during the healing phase or following reinjury. However, an important finding of the present study was that the presence of ossicles at the time of follow-up did not appear to be related to symptoms, a fact that has not been previously reported, to our knowledge. Considering the large proportion of patients with ossicles at the time of follow-up, the number of reoperations was remarkably low (two patients), indicating that the presence of ossicles at the time of follow-up is of limited relevance.
In the previous literature, excision of the ossicle(s), with or without resection of the tibial tubercle prominence, has been shown to yield better results than other methods (drilling, bone grafting, or refixation of the ossicle)9,10,18. Flowers and Bhadreshwar used tibial tubercle excision and removal of the ossicle(s) in a series of thirty-five adolescents (forty-two knees) and reported relief of pain in 95% and reduction of the prominence in 86% at the time of a five-year follow-up9. In a study with a mean duration of follow-up of 2.2 years, Orava et al. reported an excellent or good result after fifty-six of seventy operations in physically active patients who had a mean age of 19.6 years, with seventeen patients fifteen years of age or younger19. Removal of the ossicle(s) was performed in most (sixty-two) of the patients, while twenty-nine patients had other procedures, such as excision or rasping of the tibial prominence, drilling of the epiphysis, and excision of the inflamed bursa or devitalized portion of the tendon19. Because of the large number of skeletally immature patients in previous studies, comparisons of these studies with the present study of an adult population must be done with caution, particularly because the prognosis of Osgood-Schlatter disease may be somewhat better in adolescence, before the epiphysis has closed.
The radiographic analysis of the current series did not reveal anatomic variations in patellar height among patients with unresolved Osgood-Schlatter disease. The preoperative Insall-Salvati index demonstrated a normal relationship between the lengths of the patella and the patellar tendon, but, postoperatively, the tendon length was substantially increased, probably as a result of the resection of bone at the tibial tubercle. The fact that the tibial tuberosity thickness was halved after resection was also clearly documented by measuring the tibial tuberosity index. However, the reduction of the tibial tuberosity thickness did not appear to be related to the outcome of the surgery as measured with the modified Kujala score or the visual analog pain score. This observation is consistent with the findings by Orava et al.19. In addition, in the present study, neither the symptom duration before the surgery nor the surgical method (sparing or splitting of the patellar tendon) had any apparent effect on the surgical outcome; however, these findings require further study.
One of the weaknesses of the present study is that not all of the surgically treated patients were willing or able to participate in the long-term follow-up examination. Another weakness was the measurement, with the visual analog scale, of pain during the previous week, as the physical activity during that week was not taken into account. Persons with symptoms may avoid any pain-causing physical activity, resulting in bias. In addition, an obvious weakness of the study is the lack of a nonoperative treatment group.
This study has noteworthy strengths. The number of patients who participated in the long-term follow-up was much larger than the numbers in the few previous reports dealing with the outcome of surgical treatment of Osgood-Schlatter disease9-11,18,19. To our knowledge, the current study has one of the longest durations of follow-up and includes the largest number of patients with operative treatment of Osgood-Schlatter disease documented in the literature. We studied a homogeneous population of young healthy men, representing, for each study year, men of a certain age group in the entire nation. This population differs from those in most of the previous reports, which have documented the long-term results after surgical treatment of Osgood-Schlatter disease in skeletally immature adolescents9-11,29. Moreover, it was possible for us to accurately identify patients with unresolved Osgood-Schlatter disease in our study population because all military recruits were obliged to use the medical services of the military organization.
In conclusion, in this study with a median duration of follow-up of ten years, we investigated the clinical course and long-term outcome after operative treatment of unresolved Osgood-Schlatter disease in a large population of military recruits. In the great majority of patients, the surgical treatment of Osgood-Schlatter disease yielded an excellent or good functional outcome and physical activity level, and the residual pain intensity was low. In addition, one of the main symptoms, discomfort while kneeling, had totally disappeared in nearly 40% of the patients. Postoperative complications or subsequent reoperations were rare. Thus, there are no apparent long-term deleterious effects of using this technically undemanding procedure for treating this rare condition when incapacitating pain persists in young adults despite nonsurgical treatment. 