Example 1: Tibial Rotation

In this section, we will present our first example of how gait analysis and in vivo biomechanics can help quantify success in the operating room. For this example, we will focus on knee joint rotational movement patterns for which in vivo research work is scanty.

Our investigations have examined knee joint rotational movement patterns during high- and low-demand activities in both ACL deficient and reconstructed individuals. In our first study, we evaluated ACL deficient and reconstructed individuals during a low-demand activity such as walking.¹⁴ We examined 13 individuals with unilateral ACL deficiency, 21 individuals who had undergone ACL reconstruction, and 10 healthy controls. ACL reconstruction was done arthroscopically using a bone–patellar tendon–bone (BPTB) autograft. We found that the ACL deficient group exhibited significantly increased tibial rotation range of motion during the initial swing phase of the gait cycle when compared with the ACL reconstructed and control groups. Thus our results demonstrated that ACL deficiency produced rotational differences at the knee during walking. These differences did not exist when we compared the ACL reconstructed group with the control. Thus, in this low-demand activity, the surgical reconstruction restored tibial rotation to normal levels.

Next, we wanted to identify whether this is also the case in a higher-demand activity that can apply increased rotational loading at the knee. Therefore we examined 18 ACL reconstructed individuals and 15 controls during a high-demand activity (descending stairs and subsequent pivoting).²⁷ The ACL reconstruction was done arthroscopically, again using a BPTB autograft. The evaluation was performed at an average of 12 months after reconstruction. The individuals were asked to descend three steps and then immediately pivot on the landing leg at 90 degrees and walk away from the stairway while kinematic data were collected. The tibial rotation range of motion during the pivoting period was found to be significantly larger in the ACL reconstructed leg compared with the contralateral intact leg and the healthy control. No significant differences were found between the healthy control leg and the intact leg of the ACL reconstructed group. Therefore our results demonstrated that tibial rotation remained abnormal and significantly increased 1 year after ACL reconstruction during high-demand activities such as pivoting after descending from stairs.

To verify our findings, we performed an additional experiment in which we evaluated another high-demand activity.³⁰ Data were collected while the subjects jumped off a 40-cm platform and landed on the ground; following foot contact, they immediately pivoted at 90 degrees and walked away from the platform. We chose this activity because landing from a jump is a task that places higher demands on the knee than walking or even stepping down.^31,32 We combined landing with a subsequent pivoting to create rotational loads on the knee. The subjects were 11 patients, all ACL reconstructed with the same arthroscopic technique using a BPTB autograft, 1 year after the surgery; 11 ACL deficient subjects who had sustained the injury more than 1 year prior to testing; and 11 controls. The same dependent variable was evaluated as in the previous study.²⁷ Both the reconstructed leg of the ACL group and the deficient leg of the ACL deficient group had significantly larger tibial rotation values than in the healthy control group. We also found no significant differences between the deficient leg of the ACL deficient group and the reconstructed leg of the ACL reconstructed group. It was concluded that current ACL reconstruction using the BPTB autograft is inadequate to restore excessive tibial rotation during an activity such as landing and subsequent pivoting, which practically simulates sport activities.

Next, we wanted to identify whether tibial rotation remains excessive for a longer period: 2 years following the reconstruction. We speculated that it is possible adaptations will set in and the patients will compensate. Thus we performed a follow-up evaluation³³ in nine ACL reconstructed subjects who had participated in our previous study.³⁰ We examined them with the same methodology and for both activities that we used in our previous work.^27,30We also incorporated a control group of 10 individuals. We found that tibial rotation remained significantly excessive even 2 years after the reconstruction. This result was verified with comparisons conducted with both the intact contralateral knees of our patient group and with the healthy controls. Furthermore, we found that tibial rotation of the intact knee of our patient group was similar to those recorded from the healthy control group.

In all of our previous work, ACL reconstruction was performed with a BPTB autograft. Thus it was logical to question whether tibial rotation will remain excessive if an alternative autograft is used. Such an autograft is the quadrupled hamstring tendon (semitendinous and gracilis [ST/Gr]). Originally we hypothesized that the ST/Gr autograft would be able to restore tibial rotation during our experimental protocols due to its superiority in strength and linear stiffness^34–37 and because it is closer morphologically to the anatomy of the natural ACL.^34–36 We examined 11 individuals who were ACL reconstructed with an ST/Gr autograft, 11 individuals who were ACL reconstructed with a BPTB autograft, and 11 healthy controls.^38,39 The experimental protocol was identical to our previous studies. Tibial rotation was found to be significantly larger in both ACL reconstructed groups when compared with the healthy controls. Therefore our hypothesis was refuted, and we concluded that ACL reconstruction using the ST/Gr autograft is as inadequate as the one using the BPTB autograft in terms of restoring excessive tibial rotation.

The results of our studies were also supported by in vitro research work in which the biomechanical efficiency of the ACL reconstruction has also been questioned.^40–43 These studies showed that ACL reconstruction was successful in limiting anterior tibial translation in response to an anterior tibial load but was insufficient to control a combined rotatory load of internal and valgus torque. Furthermore, our tibial rotational values were in close agreement with the in vitro work.⁴⁰

In summary, our research work showed how gait analysis and in vivo biomechanics can help quantify success in the operating room. We found that ACL deficiency results in abnormal movement patterns such as excessive tibial rotation. ACL reconstruction seems to restore ACL function regarding tibial rotation in low-demand activities such as walking. However, this is not the case in higher-loading activities such as during pivoting, immediately following step-down, or in a landing from a jump. These types of activities can reveal differences that are masked during low-demand activities.

Example 2: Dynamic Functional Knee Stability Using Nonlinear Analysis

In this section we present our second example of how gait analysis and in vivo biomechanics can help measure dynamic functional knee stability. For this example, we will focus on our research work in which we used nonlinear tools to examine whether an injured joint is functionally stable during daily physical activities.

Biomechanists have recently proposed that the use of stride-to-stride variability, defined as fluctuations on the walking movement patterns from one stride to the next, provides a quantitative measure of functional joint stability.^44–47 This proposal is based on scientific evidence that neuromuscular pathology is related to an increased amount of stride-to-stride variability.^44–47 Hence a “biomechanical” hypothesis has been formed in which neuromuscular pathology is related to an increased amount of variability and deterioration of functional stability. However, this biomechanical hypothesis lacks support in other medical domains. Numerous studies in diverse medical areas have shown that a decreased amount of variability is related to pathology. These investigations include medical domains such as heart rate irregularities, sudden cardiac death syndrome, blood pressure control, brain ischemia, and epileptic seizures.^48–55 Hence a contradictory hypothesis has been proposed in which variability is described as “healthy flexibility.”^56–58 These investigations indicate that variations in the behavior of the biological system may be necessary to provide flexible adaptations to everyday stresses placed on the human body. Alternatively, a lack of healthy flexibility is associated with rigidity and inability to adapt to stresses. Based on this logic, it is possible that injury or pathology can result in a loss of healthy flexibility that may not be regained despite surgical treatment (loss of complexity hypothesis).

This contradiction in the literature may be due to the usage of linear tools (i.e., standard deviation) to assess stride-to-stride variability.^44–47 Linear tools only provide a measure of the amount of variability that is present in the gait pattern and may mask the true structure of motor variability. Masking occurs when strides are averaged to generate a “mean” picture of the subject’s gait. This averaging procedure may lose the temporal variations of the gait pattern. Additionally, the statistical processing of linear measures requires random and independent variations between subsequent strides.

Recent studies have overcome the problems of linear measures by using nonlinear tools such as the Approximate Entropy.^59–62 These studies have determined that variations in the gait pattern are distinguishable from noise and have a deterministic origin. A deterministic origin indicates that stride-to-stride variations are neither random nor independent. Rather, these variations have a meaningful pattern that characterizes the behavior of the locomotive system. Linear tools are not able to provide such information. Thus the ability to quantify the characteristic features of these variations has been the strength of using nonlinear tools to support the “loss of complexity” hypothesis.