Abstract
Background:
Postoperative residual rotatory laxity remains despite improvement in surgical techniques for anterior cruciate ligament (ACL) reconstruction (ACLR).
Purpose:
To evaluate factors associated with residual pivot shift after ACLR by quantitative measurement of the pivot shift before and after surgery.
Study Design:
Case-control study; Level of evidence, 3.
Methods:
A total of 97 patients who underwent primary double-bundle ACLR between June 2016 and March 2021 and underwent surgery to remove staples, with at least 12 months of follow-up evaluation, were enrolled. Quantitative measurements were performed under general anesthesia immediately before ACLR (preoperatively), after temporary fixation of the ACL graft (intraoperatively), and immediately before staple removal (postoperatively). The laxity of pivot shift was assessed using inertial sensors to measure acceleration and external rotational angular velocity (ERAV). Descriptive data were assessed for associations with postoperative acceleration and ERAV in a univariate analysis. A multiple linear regression analysis was performed to identify factors associated with postoperative acceleration and ERAV.
Results:
Anterior tibial translation, acceleration, and ERAV increased from intra- to postoperatively (P < .05). Factors significantly associated with postoperative acceleration were age (β = −0.238; P = .021), lateral posterior tibial slope (PTS) (β = 0.194; P = .048), and preoperative acceleration (β = 0.261; P = .008). Factors significantly affecting postoperative ERAV were age (β = −0.222; P = .029), ramp lesions (β = 0.212; P = .027), and preoperative ERAV (β = 0.323; P = .001).
Conclusion:
Greater preoperative laxity in the pivot shift was the factor having the most significant association with residual pivot shift after ACLR using quantitative measurements under general anesthesia. Younger age, higher lateral PTS, and concomitant ramp lesions were significant predictors of residual pivot shift. These findings can help pre- and intraoperative decision-making regarding whether an anterolateral structure augmentation should be added.
The pivot-shift test is a dynamic test of the rotatory laxity of the knee that produces subluxation and reduction of the lateral tibial plateau and is useful for diagnosing and managing anterior cruciate ligament (ACL) injuries. 33 Despite improvement in surgical techniques for ACL reconstruction (ACLR), postoperative residual rotatory laxity can remain, 16 as the residual positive pivot shift rate was reportedly 8% to 33%.9,19,26 Restoration of the rotatory laxity in managing ACL-injured knees is crucial because residual pivot shift is associated with worse patient-reported outcomes,4,27 increased risk of recurrent injury,28,44 and progression of osteoarthritis. 20 Several risk factors for residual pivot shift have been identified, such as younger age, 21 preoperative high-grade pivot shift,21,47 surgical delay from injury, 2 and hyperextended knees.21,47 However, the laxity of the pivot shift was assessed by subjective evaluation using grading under nonanesthesia conditions in most studies.2,47 Subjective evaluations can introduce examiner bias, 32 and nonanesthesia conditions can lead to underestimation of laxity caused by patient guarding, 8 resulting in less accurate overall assessment.
With the advancement of technology, several methods of quantitative measurement of laxity in the pivot shift have been developed for clinical use as an alternative to subjective grading.15,16 Moreover, previous studies examined the association of laxity in the pivot-shift test pre- and intraoperatively (ie, time zero) by quantitative measurements. Greater preoperative laxity in the pivot shift was reportedly a risk factor for residual instability at time zero.23,25 However, to the best of our knowledge, no studies have evaluated the factors associated with residual pivot shift at a follow-up after ACLR using quantitative measurement before and after operation under general anesthesia.
This study aimed to evaluate the factors associated with residual pivot shift after ACLR using quantitative measurement of the pivot shift under general anesthesia before and after operation. We hypothesized that greater preoperative laxity in the pivot-shift test would be a factor associated with residual pivot shift after ACLR.
Methods
Data Inclusion and Exclusion Criteria
In this retrospective study, we reviewed the records of patients with a unilateral ACL injury who underwent primary double-bundle ACLR using hamstring tendon autografts at our institution between June 2016 and March 2021. Also, the patients underwent surgery to remove staples, with at least a 12-month follow-up evaluation. We excluded patients who had recurrent or contralateral injuries after reconstruction, concomitant ligament surgery, knees with severe osteoarthritis (International Cartilage Repair Society grade 3 or 4), 29 or a positive pivot shift in the uninjured knee. Of the 287 patients initially identified, 97 were included in the final sample (Figure 1). The study protocol received ethics committee approval, and written informed consent was obtained from all included patients.

A flowchart showing the selection of patients for this study. ACL, anterior cruciate ligament.
Surgical Technique
A senior orthopaedic surgeon (M.N.) performed all ACLR procedures. The surgical technique was similar to the procedure described in a previous report. 25 The ramp lesions not identified through standard anterior portal visualization were addressed using the transintercondylar view and probing through the posteromedial capsule in all cases. 41 All ramp lesions were repaired with the knee in extension as much as possible using an all-inside suturing device (AIR, Stryker; Fast Fix 360, Smith & Nephew). The femoral tunnel was created using an inside-out approach and drilled through the medial accessory portal. After femoral fixation, temporary tibial fixation was performed for quantitative measurements of knee laxity. Each graft was fixed by clamping just above the tibial cortex using forceps under tension at 30° of flexion to <40 N for the anteromedial bundle and at 0° of flexion to <30 N for the posterolateral bundle using a ligament tensioner (Ai-Medic). After all measurements were completed, the forceps were removed, and definitive tibial fixation was performed under the same conditions as described above using double staples and bioabsorbable interference screws. All patients were informed of the removal operation before the ACLR. The removal procedure was performed only for patients who wished to undergo the surgery because of staple discomfort in the tibia, regardless of their knee joint symptoms, at a minimum 12-month follow-up after ACLR.
Quantitative Measurements of Knee Laxity
The same senior surgeon performed quantitative measurements. Measurements were performed under general anesthesia on both ACL-deficient knees and uninjured contralateral knees just before the ACLR procedure (preoperative measurement) and after temporary fixation of the ACL graft (intraoperative measurement) on the ACL-deficient knees. The postoperative measurements were performed under general anesthesia on both ACL-reconstructed knees and uninjured contralateral knees just before the operation to remove the staples.
Quantitative measurements were performed as previously described.24,25 A Rolimeter (Aircast) was used to measure anterior tibial translation (ATT; mm) during the Lachman test with maximum manual tension. To determine laxity in the pivot shift, an inertial sensor (MVP-RF8-BC; MicroStone Corp), which contains a 3-axis accelerometer and 3-axis gyroscope, was used to measure acceleration (m/s2) and external rotational angular velocity (ERAV; deg) during the pivot-shift test. An inertial sensor was fixed between the lateral aspect of the anterior tibial tuberosity and the Gerdy tubercle. The pivot-shift test was performed 9 times with an inertial sensor attached for each measurement. Data from the first and last 3 tests were excluded, and the mean of the maximum values from the middle 3 tests was calculated as the evaluation value. The pivot-shift test was performed according to a standardized pivot shift technique. 17 Petrigliano et al 36 reported that external tibial rotation integrated from ERAV during pivot shift motion was strongly correlated with potentiometer measurements of external tibial rotation. Therefore, ERAV was considered a representative value of tibial rotation during the reduction of tibial subluxation in the pivot-shift test. The high intraobserver intraclass correlation coefficient of quantitative measurements of laxity in the pivot-shift test using an inertial sensor has been reported. 31 To adjust for individual differences in the original laxity, acceleration and ERAV were represented as the side-to-side ratio from the value of the uninjured contralateral knee to that of the ACL-injured knee.
Data Collection
General data collected preoperatively were sex, age, height, body mass index, time from injury to surgery, preoperative Tegner activity score, a hyperextended knee of ≥10°, a knee extension deficit of ≥10°, and posterior tibial slope (PTS) measured on magnetic resonance imaging using a previously reported method. 18 Incidence of meniscal injury and meniscectomy were noted intraoperatively, and postoperative time from reconstruction to staple removal was recorded.
In addition, the side-to-side difference in ATT, the side-to-side ratio of acceleration, and ERAV were measured at the pre-, intra-, and postoperative time points. Finally, for subjective evaluation of rotational laxity, the pivot-shift test was performed under nonanesthesia conditions at the pre- and postoperative time points using the International Knee Documentation Committee classification (grade 0 = normal; grade 1 = glide; grade 2 = clunk; grade 3 = gross). 14
Statistical Analysis
Quantitative variables were expressed as median (interquartile range [IQR]) or frequency (percentage). Nonparametric tests, as evaluated by the Shapiro-Wilk test, were performed because most variables were not normally distributed. The Friedman test compared pre-, intra-, and postoperative knee laxity values. Bonferroni adjustment was used for the post hoc analyses. For the univariate analysis, the Mann-Whitney U test was used to compare postoperative acceleration or postoperative ERAV for the categorical variables, which were divided into 2 groups. The Spearman correlation coefficient (rS) was used to assess associations between postoperative acceleration or postoperative ERAV (as the dependent variable) and continuous variables.
A multiple linear regression analysis was performed to identify the factors associated with postoperative acceleration and postoperative ERAV. The candidate risk factors for residual pivot shift were selected according to previous studies, as follows: (1) age, 21 (2) laxity in the pivot-shift test before surgery (ie, preoperative acceleration and preoperative ERAV),21,47 (3) time from injury to surgery, 2 and (4) hyperextended knee.21,47 The candidate risk factors and pre-/intraoperative descriptive data with statistical significance in the univariate analysis were further examined using the multiple linear regression analysis and used as independent variables. For dependent variables of postoperative acceleration or postoperative ERAV, logarithmic transformation was performed before the analysis to achieve normality. Statistical significance for all tests was set at P < .05.
A post hoc power analysis was performed using G*Power Version 3.1.9.7 (Dusseldorf University). Based on the multiple regression model of acceleration and ERAV with an underlying alpha error of .01, effect sizes of 0.22 and 0.25 and power of 0.97 and 0.99 were calculated, respectively.
All statistical analyses were performed using EZR (Saitama Medical Center), 22 a graphical user interface for R Version 2.13.0 (The R Foundation for Statistical Computing). More precisely, it is a modified version of the R commander Version 1.6-3, designed to add statistical functions frequently used in biostatistics.
Results
All descriptive data for the 97 study patients and the results of the univariate analysis for postoperative acceleration and postoperative ERAV are presented in Table 1. Significant variables for postoperative acceleration in the univariate analysis were female sex, age, lateral PTS, and preoperative acceleration (P < .05). Significant variables for postoperative ERAV in the univariate analysis were sex, height, meniscal ramp lesion, and preoperative ERAV for postoperative ERAV (P < .05) (Table 1).
Descriptive Data and Results of the Univariate Analysis for Postop Acceleration and Postop ERAV a
Data are reported as n (%) or median (interquartile range). Bold P values indicate statistical significance (P < .05). Dashes indicate that a correlation analysis was not performed because of the categorical variables. ATT, anterior tibial translation; BMI, body mass index; ERAV, external rotational angular velocity; Intraop, intraoperative; Postop, postoperative; Preop, preoperative; rS, Spearman correlation coefficient.
Side-to-side difference between injured and uninjured sides.
Side-to-side ratio of injured to uninjured sides.
The multiple linear regression analysis demonstrated that the factors significantly affecting postoperative acceleration were age (P = .021), lateral PTS (P = .048), and preoperative acceleration (P = .008); those significantly affecting postoperative ERAV were age (P = .029), ramp lesions (P = .027), and preoperative ERAV (P = .001). Preoperative acceleration and preoperative ERAV were the factors with the highest standardized coefficient in each model (Table 2).
Results of Multiple Linear Regression Analysis for Risk Factors of Residual Pivot Shift a
Bold P values indicate statistically significant differences between groups (P < .05). b, unstandardized coefficient; β, standardized coefficient; ERAV, external rotational angular velocity; Postop, postoperative; Preop, preoperative; VIF, variance inflation factor.
The values obtained by logarithmically transforming the side-to-side ratio of postoperative acceleration and postoperative ERAV were used as dependent variables.
A comparison of pre-, intra-, and postoperative knee instability values over time showed that ATT, acceleration, and ERAV all significantly decreased from pre- to intraoperative (P < .05) and significantly increased from intra- to postoperative time points (P < .05) (Figure 2).

Box plots of (A) ATT, (B) acceleration, and (C) ERAV. ATT, anterior tibial translation; ERAV, external rotational angular velocity; Intraop, after temporary graft fixation; Postop, just before the operation to remove staples. Preop, just before anterior cruciate ligament reconstruction.
The results of the pivot-shift test indicated that the number of patients with grades 0, 1, 2, and 3 at preoperation were 12, 14, 61, and 10, respectively, and that the number of patients with grades 0 and 1 at postoperation were 89 and 8, respectively—the rate of residual pivot shift was 8% (8/97 patients).
Discussion
The most important finding of this study was that greater preoperative laxity of the pivot shift for acceleration and ERAV was the factor having the most significant association with postoperative laxity of the pivot shift (ie, the factor associated with residual pivot shift) after ACLR using quantitative measurements under general anesthesia. This finding validated our hypothesis. Moreover, younger age, higher lateral PTS, and concomitant ramp lesion were significant predictors of residual pivot shift. In addition, laxity in the pivot-shift test as well as that in the Lachman test increased significantly from intra- to postoperatively at a median follow-up of 17 months (IQR, 14-21 months) (Table 1).
Generally, the laxity evaluation during the pivot shift is performed using subjective grading, which leads to easy interpretation. 14 Moreover, preoperative high-grade pivot shift—mostly defined as grade ≥2—has been reported as a risk factor for residual pivot shift.21,47 However, categorization, such as grading, leads to a significant loss of information and underestimation of the variability, which reduces the statistical power and causes values that are close to each other to be categorized as a different group. 3 Therefore, the strength of our study was that continuous variables provided by quantitative measurement were used to evaluate laxity in the pivot shift both before and after surgery under general anesthesia.
There have been reports of additional anterolateral structure augmentation—such as anterolateral ligament reconstruction—to prevent residual pivot shift and lower revision rate for various indications.38,40 However, the optimal indications for additional procedures are controversial, particularly in cases of primary ACLR, because of a lack of evidence.12,42,46 Our findings enable surgeons to provide sufficient information to patients regarding the postoperative course before surgery and conduct careful follow-ups in cases with risk factors. Also, our findings can help pre- and intraoperative decision-making regarding whether an anterolateral structure augmentation should be added.
In general, the definition of “residual pivot shift” in subjective evaluation is applied 21 as grade ≥1. In the subjective evaluation under the nonanesthesia conditions of the present study, the rate of residual pivot shift was 8%, and the median side-to-side difference of ATT was 1 mm (IQR, 1-2). These results for knee stability were favorable compared with those from other reports under nonanesthesia conditions.9,14,19,26 In quantitative measurements, the definition of “residual pivot shift” has been unclear owing to the lack of evidence. However, exceeding the value on the healthy side, which is the easiest-to-understand definition, might be too strict because even side-to-side differences of ATT—which are regarded as grade A of the International Knee Documentation Committee classification—range from −1 to 2 mm, meaning that slight postoperative loosening is allowed. 14 Also, it is unclear what values are clinically acceptable in the quantitative evaluation of the pivot shift. Therefore, further research is needed to determine how postoperative values in quantitative evaluation of the pivot shift are related to clinical outcomes in the medium- and long-term follow-up.
In several reports, ramp lesions were involved in rotatory laxity.11,35,43 DePhillipo et al 11 reported that meniscocapsular and meniscotibial lesions of the posterior horn of the medial meniscus increased knee ATT, internal and external rotation, and laxity in the pivot shift in ACL-deficient knees and that the pivot shift was restored when those repairs were performed. However, our study showed that ramp lesions were 1 of the factors associated with residual pivot shift despite repairing all ramp lesions. Repairing ramp lesions using the FasT-Fix technique from the standard anterior portal, which was used in our study, has been reported and resulted in 87% complete healing on second-look arthroscopy, 10 with several advantages, such as no requirement for a posteromedial portal, easy technique, and taking less time.10,30,45 Meanwhile, Gousopoulos et al 13 reported that patients who underwent the FasT-Fix technique had a 2-fold increase in failure rate compared with those who underwent suture hook repair at a mean follow-up of 97.7 months. Gousopoulos et al explained that the injured meniscotibial ligament retracts inferiorly, making it difficult to reach the meniscotibial ligament with the standard anterior approach. However, whether healing of the meniscotibial ligament was achieved was not determined on second-look arthroscopy in their report. Furthermore, suture hook repair has several disadvantages, such as the risk of saphenous nerve and vein injury 45 and the need for advanced surgical skills. 1 Therefore, the appropriate repair method for ramp lesions and the cause of residual laxity should be further examined.
A higher PTS has been considered a risk factor for high-grade pivot shift, 39 ACL rupture, 7 and ACLR failure. 34 In particular, the lateral tibial slope, a factor associated with residual pivot shift in our study, was reportedly related to high-grade rotatory laxity.5,37 In a previous study 25 regarding time zero evaluation with quantitative measurement of the pivot shift, the PTS was not a risk factor for residual pivot shift. The discrepancy between this finding and ours may be caused by the duration of the graft fixation to laxity evaluation. A steeper slope in biomechanical laboratory studies with cadaveric knees increased ACL graft force under loading with the tibiofemoral compression force, anterior force, valgus moment, and combined conditions.6,48 This finding suggested that the ACL of the knee with a steeper slope is subjected to greater tension during the postoperative follow-up. Therefore, a steeper tibial slope can contribute to more subsequent graft elongation and knee laxity during postoperation than intraoperation.
Limitations
This study has several limitations. First, this was a retrospective study. In the latter period of the study, additional anterolateral ligament reconstruction was performed in some cases with a hyperextended knee, chronic injury, high activity, and gross pivot shift at the time of preoperative examination, which was excluded. This might explain the absence of association between hyperextended knees and chronic cases with residual pivot shift. Second, because only patients who underwent removal surgery were included, there might have been a selection bias against patients who did not undergo removal surgery. Third, our study results were evaluated only in the short-term postoperative period, and clinical outcomes—such as patient-reported outcomes—were not examined. However, most reports found a relationship between the laxity of the pivot shift after surgery and clinical outcomes,4,27 risk of recurrent injury,28,44 and progression of osteoarthritis. 20 Therefore, our results are meaningful for the treatment of ACL injuries. Fourth, other factors that may be involved with postoperative laxity of the knee—such as generalized ligamentous laxity, postoperative activity level, the period from surgery to return to sports, and anterolateral structure injuries—were not evaluated. Fifth, temporary fixation in intraoperative measurements was not exactly the same as that of definitive fixation and did not fully reproduce the knee instability after definitive fixation. The distance of fixation points in the definitive fixation is slightly decreased compared with that of temporary fixation. Therefore, instability may be reduced if measurements are performed immediately after the definitive fixation; nonetheless, further study is needed to prove this.
Conclusion
In this study, greater preoperative laxity in the pivot shift was the factor having the most significant association with residual pivot shift after ACLR using quantitative measurements under general anesthesia. Moreover, younger age, higher lateral PTS, and concomitant ramp lesions were significant predictors of residual pivot shift. These findings enable surgeons to provide sufficient information about the postoperative course to patients before surgery and conduct careful follow-ups in cases with risk factors but also help pre- and intraoperative decision-making regarding whether an anterolateral structure augmentation should be added.
Footnotes
Final revision submitted August 16, 2023; accepted August 29, 2023.
The authors have declared that there are no conflicts of interest in the authorship and publication of this contribution. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
Ethical approval for this study was obtained from Nagoya City University (60-18-0154).
