Abstract
Purpose
The Forgotten Joint Score (FJS-12) is a scoring system initially created to assess post arthroplasty outcomes. It has since been used to evaluate Anterior Cruciate Ligament surgery outcomes. Our study aims to evaluate the applicability and validity of the FJS-12 in post Anterior Cruciate Ligament reconstructed patients, and to assess correlation with established Patient Reported Outcome Measure Scores in the same population.
Design
Case series, level 3 evidence.
Methods
We conducted a cross sectional study across patients who had undergone Anterior Cruciate Ligament reconstruction and carried out the FJS-12 questionnaire by phone interview. Patients who had undergone primary Anterior Cruciate Ligament reconstruction were considered for the study.
Results
The average Forgotten Joint Score-12 for all 82 patients was 71.4 (±22.9), which corresponded to a normal distribution. The average Lysholm and Tegner score at the 2-year post-operative visit was 93.5 ± 9.5 and 5.8 ± 1.8 respectively and the distribution was non-normal. We noticed a large ceiling effect of 42.7% in the Lysholm scores, but only 8.4% in FJS-12. There was a weak correlation with Lysholm and a positive correlation with Tegner.
Conclusions
Forgotten Joint Score-12 seems to be a promising patient reported outcome measure that can be used in evaluating post Anterior Cruciate Ligament reconstruction outcomes. It is more discerning than traditional scores and is easy to administer thus it can used in the clinical follow-up of patients. With the scores being normally distributed, it makes for a meaningful PROMS and would allow more accurate application of parametric statistical tests.
Keywords: Forgotten joint score, Patient reported outcome, General sports trauma, Knee, Ligaments, Anterior cruciate ligament reconstruction
Abbreviations: ACL, Anterior Cruciate Ligament; FJS-12, Forgotten Joint Score - 12 questions; PROMS, Patient Reported Outcome Measure Scores
1. Introduction
The Forgotten Joint Score (FJS-12) created in 2013, has since been validated in various international studies for evaluating post arthroplasty outcomes. It has been praised for its low ceiling effect and the ability to capture more subtle data.1, 2, 3 Achieving a “forgotten knee” has become the goal of many arthroplasty centres adopting the use of the FJS-12 and is a concept that is increasingly being applied to other aspects of orthopaedic surgery. The original creators of the FJS-12 Behrend et al. have evaluated the use of FJS-12 in ACL reconstructed knees in 2017, comparing the score to Knee Injury and Osteoarthritis Outcome Score (KOOS) and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores in medium- and long-term outcomes.4 In a second study the authors compared post ACL reconstructed knees to healthy controls and obtained reference values, finding the FJS-12 to be able to differentiate significantly between the two groups.5
As ACL reconstruction outcomes improve and patient expectations increase over the years, traditional outcome assessment tools seem to be losing their discriminatory abilities to differentiate outcomes for high functioning athletes.6, 7, 8 Studies have found no significant statistical difference for many of the different new techniques that seem to have better biomechanical properties, leaving researchers and clinicians in continuous debate.9,10
The aim of our study was to assess the applicability and validity of the FJS-12 score in evaluating post ACL reconstructed patients in our local population and correlate it with established objective Patient-Reported Outcome Measures (PROMs). The clinical relevance being that FJS-12 would be a much shorter and easier outcome parameter to measure compared to traditional instruments, and if validated in the local population, would be a useful tool to monitor outcomes for different ACL reconstruction techniques.
2. Methods
Institutional Review Board approval was obtained under the local Centralized Institutional Review Board with reference number 2017/2240 which agreed with the ethical standards of the Singhealth Integrated System in Healthcare for Research committee and with the 1964 Helsinki declaration. All patients that had undergone ACL reconstruction at our tertiary institution and operated by either one of our 2 fellowship trained senior authors between 2012 and 2016 were considered for enrolment in the study. Inclusion criteria were: (1) minimum follow-up two years post operation with Tegner and Lysholm scores, (2) unilateral ACL rupture; (3) primary ACL reconstruction with hamstring autograft, trans portal approach anatomic ACL reconstruction with endobutton fixation device, (4) informed consent. We included patients under these two surgeons as they used similar methods of ACL reconstruction and shared the same protocols.
Exclusion criteria were
-
•
Multiple ligamentous injuries
-
•
Injuries with associated fractures of the lower extremity
-
•
Previous unrelated surgeries to the knee
-
•
Revision ACL reconstruction surgery
-
•
Post ACL reconstruction graft failure or ligamentous reinjures
Suitable patients were recalled, and informed consent was obtained before carrying out the FJS questionnaire (FJS-12).
2.1. ACL reconstruction technique
The fellowship trained senior authors have more than 15 years of experience in anatomic ACL reconstruction. Both surgeons used a trans-portal technique to reconstruct the ACL based on the footprint. Where possible both hamstrings are quadrupled and attached to an endobutton device for double bundle reconstruction. In the case that one bundle is seen to be intact, it was preserved and only a selective bundle reconstruction was performed.
2.2. Rehabilitation protocol
All patients followed the same rehabilitation protocol with early active quadriceps isometric exercises and active ROM exercises. Patients were allowed to fully weight-bear on the first postoperative day in a knee brace with a range from 0 to 90°. This was gradually weaned off at 4 weeks post operation. Patients were encouraged to start cycling at 3 months and straight line running at 6 months after surgery. Ability to return to full sporting activity with directional changes was assessed at 1 year after surgery by qualified physiotherapists, and patients were given individualized goals thereafter. At 2 years post operatively, Lysholm and Tegner scores were assessed by a team of independent assessors.
2.3. Forgotten Joint Score-12
The FJS-12 total score is derived from the sum of the individual responses that is linearly transformed to a metric ranging from 0 to 100 points.3 High scores indicate a high level of forgetting the joint, i.e. low joint awareness. Being in a multinational country, there was one situation where we had to use the Simplified Chinese version of the FJS-12.11 The other 81 patients were able to understand and answer the questionnaire in the English language.
2.4. Statistical analysis
Statistical analysis was performed using Statistical Package for the Social Sciences (IBM SPSS® software platform). Cronbach's alpha was calculated to study internal consistency of FJS-12. We investigated correlation between score systems using Spearman's correlation coefficient and inter correlation of FJS-12 using inter-item correlation and correlations between item and total score.
The Kruskal-Wallis one-way analysis of variance was used to determine if there were statistically significant difference in patients’ demography and reconstruction outcomes between Single bundle, Double bundle and Selective bundle ACL. All tests of significance were two-tailed with p-value calculations 0.05.
A priori power analysis (80% power, alpha = 0.01) using the same FJS-12 statistics as Behrend's paper (estimated mean = 90, estimated SD13) suggested that our sample should contain at least 24 cases (Cohen's d 1.00).
3. Results
Eighty-two responses were recorded from eighty-two patients who were aged from 15.3 to 55.5 years old. The average age of our patients at operation was 27.7 years old and six patients were above 40 years of age. Body Mass index 25.5 ± 3.7 kg/m2. There were sixteen Single bundle, twenty-seven Selective bundle and thirty-nine Double bundle ACL reconstructions. Years from surgery was on average 4.9 with a range from 3.0 to 7.0 years.
The average Lysholm score attained was 93.5 ± 9.52 with a ceiling effect of 42.7%. The Lysholm score for the eighty-two patients ranged from 49 to 100 with a mode of 100. The average Tegner score at the same visit was 5.8 ± 1.8 with a ceiling effect of 2.4% and a mode of 7. On the Tegner activity scale it meant the average patient was able to do recreational sports and some competitive sports between two to five times weekly. The scores ranged from 2 to 10 and thus eighty-two patients had a total of 9 different scores. The mean, standard deviation and plots of the total scores are compared in a box and whiskers plot (Fig. 1).
Fig. 1.
Box and Whiskers plot of the scores.
The mean FJS-12 score for our eighty-two patients was 71.4 (±22.9) with a median of 77.1.8.5% of patients scored the maximum score of 100% which also happened to be the mode. Eighty-two patients had thirty-three different values for the FJS-12 final score, with a range from 14.5 to 100. We found excellent internal consistency for all 12 questions with a Chronbachs’ α of 0.9 The average inter-item correlation for the questions are positive at 0.43 and the corrected total item correlation was 0.68.
We used both the Pearson's and the Spearman's correlation coefficient to compare FJS-12 to Tegner and Lysholm scores (Fig. 3). Both the correlation coefficients showed similar findings of poor correlation between Lysholm and FJS-12 and a moderately positive correlation between Tegner and FJS-12.
Fig. 3.
Correlation of FJS-12 to Tegner and Lysholm scores.
The FJS-12 scores in our sample had a Kolmogorov-Smirnov test statistic of 0.11 (p-value 0.21) with a skewness of −0.68 and a kurtosis of −0.50. Thus, the data was found not to differ from one that is normally distributed. For the Lysholm scores of the patients, we calculated the Kolmogorov-Smirnov test statistic to be 0.26 (P < 0.01) with skewness of −2.25 and a kurtosis of 6.08. For the Tegner scores of the patients, we calculated the Kolmogorov-Smirnov test statistic to be 0.16 (P = 0.03) with skewness of −0.08 and a kurtosis of −0.50. Thus, the Lysholm and Tegner scores were not normally distributed (Fig. 2). Using the two-tailed student's T-test, there was no significant difference between the scores of the 6 participants aged 40 and above from the rest of the participants (P = 0.71). The Anova one-way analysis of variance was used to compare between Single bundle, Double bundle and Selective bundle ACL reconstruction outcomes. Between the different reconstruction techniques, the FJS-12 elicited a larger difference in outcomes whereas Tegner and Lysholm scores did not. Even though P value did not reach level of significance we found it clinically relevant. Further analysis of the FJS-12 scores was carried out with the Man-Whitney U Test which showed Selective bundle ACL reconstruction resulting in significantly better outcomes than Single bundle ACL reconstruction (p = 0.02).
Fig. 2.
Frequency of scores in distribution curve.
4. Discussion
The most important finding of the present study was that FJS-12 scores showed a normal distribution, which corresponds to the Central Limit Theorem. None of the other scoring systems for ACL reconstruction have data that is normally distributed, even when applied to a population with normal knees.12,13 Most biological variables follow a normal distribution or a log-normal curve, and it is ideal if the outcome measure can reflect this.14 In clinical research studies, a normalized data set is preferred for application of standard parametric tests to analyse scores with better accuracy.15,16
The FJS-12 had better discriminatory power as compared to Lysholm and Tegner in terms of score distribution, ceiling effect and during the comparison of ACL reconstruction techniques. The FJS-12 did not correlate with Lysholm scores and showed moderate correlation with Tegner scores. The high ceiling effect of the Lysholm scores and its skewed distribution towards upper ranges of the score could have contributed to this. This large ceiling effect seen in Lysholm scores was also noted in older studies.17 The FJS may have correlated better with the Tegner activity scale as patients could have been more willing to participate in activities when their affected joint does not bother them. However, the majority of local sportsmen participate in recreational sports and thus may not score high on the Tegner scale, despite having excellent outcomes. Because sports goals are individual to each patient, we felt that the FJS-12 was inclusive to all patients regardless of their competitive level. Furthermore, Tegner has limited data plots whereas the FJS-12 was able to capture more variation in scores. Therefore, the FJS-12 may be more applicable to the current generation of ACL injured athletes.
While conducting interviews, we noted some difficulties with specific questions. Question 5 regarding knee awareness while travelling in a car was vague and could be interpreted as either being the driver or the passenger. Question 4 regarding knee awareness while taking a bath was thought to yield variable answers according to type of showering facilities. Notably in our population this was also the question that had the highest number of good scores with 87.8% scoring the highest level of forgetting their joint, which could be due to our culture of housing having mostly standing showering stalls. Omitting this question resulted in an improved Cronbach's alpha.
4.1. Limitations
We were not able to collect FJS scores in the preoperative and rehabilitation period and were unable to assess the ability of the FJS-12 to measure the clinically meaningful difference and sensitivity to change over time.
4.2. Conclusion
We conclude that the FJS-12 is a valid patient reported outcome score to assess outcomes of ACL reconstruction surgery with low ceiling effect and high internal consistency. It is concise, easy to administer, and is applicable to a wide range of patients. The scores being normally distributed follow the ideal biological model and standard parametric tests can be applied with greater accuracy. It shows better discerning capabilities as compared to Lysholm and Tegner scores and could be the solution to finding the significance in comparative clinical studies for ACL reconstruction techniques. More studies are needed to compare pre-operative scores and trend the FJS-12 scores over time and measure effect sizes.
Funding
The authors state that Grants, financial support and technical or other assistance were not used in the preparation of this research paper. This research project did not receive funding or external financial support, and we have no conflicts of interest to disclose nor competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Author statement
Jia Ying Lee was the main author in charge of analysing and writing the manuscript. Yi Mei Low and Ying Hao assisted in the statistical analysis. Zi Yang Chia and Lei Jiang did the conceptualizing of the article, and Senior authors Denny Lie and Paul Chang were the surgeons whose patients were the participants. They also advised on the direction of the manuscript and did the final edits. The authors state that Grants, financial support and technical or other assistance were not used in the preparation of this research paper. This research project did not receive funding or external financial support, and we have no conflicts of interest to disclose nor competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
There were no grants, financial support or technical assistance that supported this research project. The authors have no Conflicts of Interest to declare
References
- 1.Hamilton D.F., Loth F.L., Giesinger J.M. Validation of the English language Forgotten Joint Score-12 as an outcome measure for total hip and knee arthroplasty in a British population. Bone Jt J. 2017;99-B(2):218–224. doi: 10.1302/0301-620X.99B2.BJJ-2016-0606.R1. [DOI] [PubMed] [Google Scholar]
- 2.Robinson P.G., Rankin C.S., Lavery J., Anthony I., Blyth M., Jones B. The validity and reliability of the modified forgotten joint score. J Orthop. 2018;15(2):480–485. doi: 10.1016/j.jor.2018.03.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Behrend H., Giesinger K., Giesinger J.M., Kuster M.S. The “forgotten joint” as the ultimate goal in joint arthroplasty. Validation of a new patient-reported outcome measure. J Arthroplasty. 2012;27(3):430–436. doi: 10.1016/j.arth.2011.06.035. e1. [DOI] [PubMed] [Google Scholar]
- 4.Behrend H., Giesinger K., Zdravkovic V., Giesinger J.M. Validating the forgotten joint score-12 in patients after ACL reconstruction. Knee. 2017;24(4):768–774. doi: 10.1016/j.knee.2017.05.007. [DOI] [PubMed] [Google Scholar]
- 5.Behrend H., Zdravkovic V., Giesinger J.M., Giesinger K. Joint awareness after ACL reconstruction: patient-reported outcomes measured with the Forgotten Joint Score-12. Knee Surg Sport Traumatol Arthrosc. 2017;25(5):1454–1460. doi: 10.1007/s00167-016-4357-x. [DOI] [PubMed] [Google Scholar]
- 6.Mohtadi N.G., Chan D.S. A randomized clinical trial comparing patellar tendon, hamstring tendon, and double-bundle ACL reconstructions: patient-reported and clinical outcomes at 5-year follow-up. J Bone Jt Surg Am. 2019;101(11):949–960. doi: 10.2106/JBJS.18.01322. [DOI] [PubMed] [Google Scholar]
- 7.Tie K., Chen L., Hu D., Wang H. The difference in clinical outcome of single-bundle anterior cruciate ligament reconstructions with and without remnant preservation: a meta-analysis. Knee. 2016;23(4):566–574. doi: 10.1016/j.knee.2015.07.010. [DOI] [PubMed] [Google Scholar]
- 8.Chirurg D. 2019. Vergleichbare Ergebnisse nach arthroskopischem Ersatz des vorderen Kreuzbandes Klinische und funktionelle Ergebnisse nach. [DOI] [PubMed] [Google Scholar]
- 9.Yoon K.H., Kim J.S., Kim S.J., Park M., Park S.Y., Park S.E. Eight-year results of transtibial nonanatomic single-bundle versus double-bundle anterior cruciate ligament reconstruction: clinical, radiologic outcomes and survivorship. J Orthop Surg. 2019;27(2):1–8. doi: 10.1177/2309499019840827. [DOI] [PubMed] [Google Scholar]
- 10.Dong Z., Niu Y., Qi J., Song Y., Wang F. Long term results after double and single bundle ACL reconstruction: is there any difference? A meta - analysis of randomized controlled trials. Acta Orthop Traumatol Turcica. 2019;53(2):92–99. doi: 10.1016/j.aott.2018.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Cao S., Liu N., Han W. Simplified Chinese version of the Forgotten Joint Score (FJS) for patients who underwent joint arthroplasty: cross-cultural adaptation and validation. J Orthop Surg Res. 2017;12(1):1–7. doi: 10.1186/s13018-016-0508-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Briggs K.K., Lysholm J., Tegner Y., Rodkey W.G., Kocher M.S., Steadman J.R. The reliability, validity, and responsiveness of the lysholm score and tegner activity scale for anterior cruciate ligament injuries of the knee: 25 years later. Am J Sports Med. 2009;37(5):890–897. doi: 10.1177/0363546508330143. [DOI] [PubMed] [Google Scholar]
- 13.Briggs K.K., Steadman J.R., Hay C.J., Hines S.L. Lysholm score and tegner activity level in individuals with normal knees. Am J Sports Med. 2009;37(5):898–901. doi: 10.1177/0363546508330149. [DOI] [PubMed] [Google Scholar]
- 14.Koch A.L. The logarithm in biology 1. Mechanisms generating the log-normal distribution exactly. J Theor Biol. 1966;12(2):276–290. doi: 10.1016/0022-5193(66)90119-6. [DOI] [PubMed] [Google Scholar]
- 15.Maltenfort M. Understanding a normal distribution of data. J Spinal Disord Tech. 2015;28(1):377–378. doi: 10.1097/BSD.0000000000000337. [DOI] [PubMed] [Google Scholar]
- 16.Maltenfort M. Understanding a normal distribution of data (Part 2) J Spinal Disord Tech. 2016;29(1):30. doi: 10.1097/BSD.0000000000000337. [DOI] [PubMed] [Google Scholar]
- 17.Ra H.J., Kim H.S., Choi J.Y., Ha J.K., Kim J.Y., Kim J.G. Comparison of the ceiling effect in the Lysholm score and the IKDC subjective score for assessing functional outcome after ACL reconstruction. Knee. 2014;21(5):906–910. doi: 10.1016/j.knee.2014.06.004. [DOI] [PubMed] [Google Scholar]