Influence of Severity and Duration of Anterior Knee Pain on Quadriceps Function and Self-Reported Function

Initially, 214 volunteers aged 18 to 35 years were recruited from the university and local community via flyers and screened for eligibility, resulting in 60 patients with AKP and 48 healthy control individuals whose data were used in the final analysis (Table 1; Figure). To minimize potential confounding factors, healthy control individuals were matched on age, height, mass, body mass index, thigh circumference, and physical activity (Tables 2–5). All participants had to engage in ≥150 minutes of moderate exercise (eg, walking and light jogging) or 60 minutes of vigorous exercise (eg, running and weightlifting) per week.²⁶  After taking measurements, we categorized patients into 3 subdivisions based on pain (1) severity (low [≤3 of 10] versus high [>3 of 10]), (2) duration (short [<2 years] versus long [>2 years]), and (3) severity and duration (low and short [≤3 of 10, <2 years] versus low and long [≤3 of 10, >2 years] versus high and short [>3 of 10, <2 years] versus high and long [>3 of 10, >2 years]). The university's institutional review board approved the study.

Upon arrival at the laboratory, participants read and gave informed consent. After anthropometric assessment using a bioelectrical impedance device (model InBody 770; Biospace Ltd), health history questionnaires were completed to collect the history of injuries and surgeries, physical activity, severity of typical pain in the past week²⁸  using the NPRS (0 = no pain, 10 = worst pain imaginable), and pain duration. We used the Lower Extremity Functional Scale (LEFS) to evaluate self-reported function.²⁹  Then thigh circumference (at the midpoint between the anterior-superior iliac spine and the superior pole of the patella) in centimeters was obtained using a tape measure.

After a 10-minute warm-up exercise on a stationary bike at a self-selected pace, the participant was seated on an isokinetic dynamometer (model 770 Norm; Cybex International). The knee and hip joints were positioned at 90° and 85°, respectively. The chest, pelvis, and thigh on the nonpainful side were secured with straps to prevent accessory motions or assistant muscle contractions. Two 7- × 12.7-cm stimulating electrodes (model Dura-Stick II; Chattanooga Group Inc) were attached to the proximal vastus lateralis and distal vastus medialis after the anterior thigh was shaved with a razor and cleaned with alcohol swabs. For the matched healthy control group, the dominant leg (defined as the leg used to kick a ball) was evaluated.

After familiarization sessions,²⁶  participants performed 3 trials of quadriceps MVIC with 90-second rest intervals. Visual feedback and oral encouragement (“as fast and hard as possible”) were provided for each trial. When knee-extension torque reached a plateau, an electrical stimulus (100 pulses/s, 600-microsecond pulse duration, a 100-millisecond train of 10 stimuli, 125 V with a peak output current of 450 mA) was delivered to the quadriceps via 2 stimulating electrodes to recruit any remaining motor units.¹¹  We used the S48 Grass Stimulator with an SIU8T transformer stimulus isolation unit (Grass-Telefactor Inc) to generate the superimposed burst (SIB). After a 5-minute rest, quadriceps endurance was concentrically measured under isokinetic conditions at 180°/s with 20 repetitions.³⁰  The range of motion for the isokinetic condition was set from 90° to 20° of knee flexion.²¹ 

Using isometric measurements, we averaged 3 trials of quadriceps MVIC and normalized to body mass (N·m/kg). The quadriceps CAR was calculated by F_MVIC/F_MVIC+SIB (ratio).¹²  The quadriceps RTD was determined as the average slope (Δtorque/Δtime) of the torque-time curve over the first 200 milliseconds and normalized to body mass and time interval (N·m/s/kg).³¹  Using isokinetic measurements, 1 trial of quadriceps APT that contained 20 repetitions was recorded, and the calculated mean value was normalized to body mass (N·m/kg).³⁰ 

Our sample size was determined using an expected mean difference in CAR of 0.11, with an SD of 0.14.¹¹  With α = .05 and β = 0.2, we estimated that 14 participants in each group would be necessary.

Means and 95% CIs were calculated. Before participants were classified into subgroups by severity and duration of AKP, independent t tests were performed to compare participant demographics (except for sex), pain severity and duration, quadriceps function, and self-reported function. A 1-way analysis of variance and Tukey-Kramer post hoc tests were conducted when patients with AKP were categorized 3 times into either 2 or 4 subgroups (Figure) based on severity (≤3 of 10 versus >3 of 10 on NPRS)^23,26  and duration (<2 years versus >2 years)^24,25  of AKP and compared with their healthy counterparts. To assess practical significance, Cohen d effect sizes ([₁ − ₂]/σ_pooled) with 95% CIs were also computed. Pearson r correlation coefficients were calculated to determine the bivariate correlations among outcome measurements. The statistical package SAS (version 9.4; SAS Institute) was used for all tests (P < .05).

Compared with the healthy control group, participant demographics were not different in the AKP group (P values ≥ .13; Table 2) or any AKP subgroups (P values ≥ .08; Tables 3–5). The AKP subgroup with a high severity of pain showed greater mass (P = .04; Table 3) and body mass index (P = .04; Table 3) than the AKP subgroup with a low severity of pain.

Versus the healthy control group, the AKP group (P < .0001, d [95% CI] = 3.30 [2.72, 3.88]; Table 2) and all AKP subgroups (P < .0001, d [95% CI] ≥ 3.77 [2.93, 19.95]; Tables 3–5) showed more severe symptoms of pain. All AKP subgroups with a high severity of pain demonstrated more severe symptoms of pain compared with the AKP subgroups with a low severity of pain (P < .0001, d [95% CI] ≥ 3.67 [2.46, 5.27]; Tables 3 and 5).

Compared with the healthy control group, the AKP group (P < .0001, d [95% CI] = 1.58 [1.15, 2.01]; Table 2) and AKP subgroups (P < .0001, d [95% CI] ≥ 1.87 [1.26, 5.97]; Tables 3–5) exhibited longer symptoms of pain, except for the AKP subgroups with low severity and short duration (P = .06; Table 5) and high severity and short duration (P = .08; Table 5) of pain. All AKP subgroups with a long duration of pain displayed longer symptoms of pain than the AKP subgroups with a short duration of pain (P < .0001, d [95% CI] ≥ 2.49 [1.52, 3.72]; Tables 4 and 5).

Versus the healthy control group, (1) the AKP group and AKP subgroups had less quadriceps maximal strength (AKP group: P < .0001, d [95% CI] = 1.06 [0.65, 1.47]; Table 2; AKP subgroups: P < .005, d [95% CI] ≥ 0.78 [0.25, 2.33]; Tables 3–5) and activation (AKP group: P < .0001, d [95% CI] = 0.94 [0.54, 1.34]; Table 2; AKP subgroups: P < .02, d [95% CI] ≥ 0.85 [0.32, 2.29]; Tables 3–5), except for the AKP subgroup with a low severity and short duration of pain (P > .32; Table 5); and (2) the AKP group and AKP subgroups with a high severity or long duration of pain showed less quadriceps explosive strength (AKP group: P = .004, d [95% CI] = 0.56 [0.17, 0.95]; Table 2; AKP subgroups: P < .007, d [95% CI] ≥ 0.74 [0.26, 1.27]; Tables 3 and 4) and endurance (AKP group: P = .001, d [95% CI] = 0.66 [0.27, 1.05]; Table 2; AKP subgroups: P < .003, d [95% CI] ≥ 0.79 [0.31, 1.42]; Tables 3 and 4). When severity and duration were combined, however, only the AKP subgroup with high severity and long duration of pain had less quadriceps explosive strength (P = .006, d [95% CI] = 1.09 [0.51, 1.67]; Table 5) and endurance (P = .0004, d [95% CI] = 1.21 [0.62, 1.80]; Table 5).

Compared with the healthy control group, the AKP group (P < .0001, d [95% CI] = 2.99 [2.44, 3.54]; Table 2) and all AKP subgroups (P < .0001, d [95% CI] ≥ 3.44 [2.66, 7.25]; Tables 3–5) demonstrated less self-reported function. All AKP subgroups with a high severity of pain displayed less self-reported function than the AKP subgroups with a low severity of pain (P < .005, d [95% CI] ≥ 0.92 [0.19, 1.93]; Tables 3 and 5).

Pain severity and duration were positively correlated (r = 0.53; Table 6). Both pain severity and duration were negatively correlated with quadriceps function and self-reported function (r ranged between −0.85 and −0.28; Table 6). All quadriceps function and self-reported function values were positively correlated (r ranged between 0.20 and 0.56; Table 6).

Our purpose was to examine how each combination of severity and duration of AKP influenced quadriceps function and self-reported function. Our hypotheses were accepted: both severity and duration of AKP were inversely associated with quadriceps function and self-reported function. Also, a combination of high severity and long duration of AKP caused further deficits in quadriceps function.

To our knowledge, this is the first study to categorize patients with AKP into subdivisions based on pain severity (low [≤3 of 10] versus high [>3 of 10])^23,26  and duration (short [<2 years] versus long [>2 years])^24,25  and compare them with a healthy population. Except for the AKP subgroup with a low severity and short duration of pain (Table 5), all AKP subgroups exhibited deficits in quadriceps maximal strength (MVIC) and activation (CAR; Tables 3–5). This indicates that a combination of lower perceived (eg, ≤3 of 10) and shorter-lasting (eg, <2 years) AKP may not sufficiently contribute to quadriceps neuromuscular changes. For example, using an experimental knee pain model, investigators³²  demonstrated that pain severity of 3.2 of 10 caused quadriceps weakness, and this weakness was positively associated with pain severity. As an inverse approach, another experimental study³³  using a reduced knee pain model revealed that quadriceps strength was restored when pain severity decreased from 3.8 to 1.5 and from 3.9 to 2.2 after local anesthetic and placebo interventions, respectively. This evidence^32,33  may imply that pain beyond a certain severity level (eg, <3 of 10) is necessary for a short duration of AKP to induce quadriceps dysfunction. This is also supported by our data reflecting that the AKP subgroup with a high severity and short duration of pain (5.7 of 10 and 11.3 months, respectively; Table 5) showed deficits in quadriceps maximal strength and activation, despite having a similar pain duration as the AKP subgroup with a low severity and short duration of pain (2.6 of 10 and 11.4 months, respectively; Table 5). Yet future longitudinal research is needed to identify the intensity of AKP required to cause muscle dysfunction at any specific time point before it develops into chronic AKP. In terms of pain duration, the authors⁷  of a previous study hypothesized that the degree of neural adaptation in the central nervous system due to pain is potentially influenced over time by persistent peripheral nociceptor activity. Thus, impaired muscle strength of the lower extremity could be a result of persistent AKP.^10,21  We also found that the AKP subgroup with a low severity and long duration of pain (2.5 of 10 and 58.0 months, respectively; Table 5) had less quadriceps maximal strength and activation, despite having a similar pain severity as the AKP subgroup with a low severity and short duration of pain (2.6 of 10 and 11.4 months, respectively; Table 5). Finally, all these outcomes highlight the importance of pain control in the early phase of rehabilitation to prevent or delay the progression of subsequent quadriceps impairments.

Deficits in quadriceps explosive strength (RTD) and endurance (APT) were apparent in the AKP subgroups with a high severity (Table 3) or long duration (Table 4) of pain, but when severity and duration were combined, only the AKP subgroup with a high severity and long duration of pain showed less quadriceps explosive strength and endurance (Table 5). This may indicate that additional quadriceps dysfunction is likely in patients with more severe and longer-lasting AKP. Traditionally, a general guideline for a successful return to activity is to achieve approximately 85% to 90% of the maximal strength relative to the asymptomatic limb³⁴ ; however, evaluation of quadriceps function in various contraction modes (eg, explosive strength and endurance) that are more relevant to biomechanical factors and functional movements should be considered in clinical practice.^31,35  Quadriceps maximal strength (MVIC) was restored in 6 months (97%), whereas quadriceps explosive strength (RTD) took a longer period (ie, 12 months) to reach 90% of the preinjury level.³⁶  Additionally, explosive strength³⁶  and endurance³⁵  of the quadriceps were more sensitive to alterations in neuromuscular function and joint loading from a long-term perspective than isometric strength. Thus, we suggest that clinicians incorporate and analyze various characteristics of the muscle-force waveforms (eg, RTD and APT) and use this information in their clinical practice along with the MVIC to identify the specific muscle capacity to be further targeted and treated. Greater deficits in quadriceps explosive strength and endurance resulting from a high severity and long duration of pain would precipitate early onset and progression of degeneration in the knee joint.^31,35  Patients with chronic unilateral AKP (4.4 of 10 and 48.6 months, respectively) in earlier data²¹  had bilateral deficits in quadriceps maximal and explosive strength, activation, and endurance, despite having a lower pain severity and shorter pain duration than the AKP subgroup with a high severity and long duration of pain (5.2 of 10 and 67.1 months, respectively; Table 5). This evidence may also support the idea that our current AKP subgroup with a high severity and long duration of pain should be evaluated and treated bilaterally.

Compared with the healthy control group, all AKP subgroups showed lower scores in self-reported function (Tables 3–5). Interestingly, the AKP subgroup with a low severity and short duration of pain did not experience any quadriceps dysfunction but had subjective functional limitations (Table 5). As the authors³⁷  of a previous study observed, this suggests that the perception of pain could independently exacerbate self-reported function without muscle dysfunction. Regardless of pain duration, the AKP subgroups with a high severity of pain expressed less self-reported function than the AKP subgroups with a low severity of pain (Tables 3 and 5). A direct association between pain severity and self-reported function (LEFS) in this population has been noted.²⁶  Our data also indicated that pain duration was moderately correlated (r = −0.56) with self-reported function, whereas pain severity had a strong correlation (r = −0.85; Table 6). Earlier investigators²⁷  determined that a rehabilitation program accompanied by patterned electrical neuromuscular stimulation (50-Hz pulse frequency, 70-microsecond phase duration, and 200-millisecond stimulus train on the gluteus medius, vastus medialis oblique, hamstrings, and adductors for 15 minutes before therapeutic exercise for 3 sessions over 4 weeks) reduced pain and improved self-reported function despite chronic AKP (>2 years). The clinical implication is that more attention should be paid to pain intensity, which is a controllable factor in clinical practice.

The relationships among pain severity and duration, quadriceps function, and self-reported function were characterized (Table 6). Pain severity and duration were moderately positively correlated (r = 0.53), in agreement with previous findings.¹⁶  Because a cause-and-effect relationship cannot be assessed through the current (cross-sectional) study design, future researchers should evaluate whether an increase in pain leads to persistent pain or, conversely, whether persistent pain leads to an increase in pain. This may provide a deeper understanding of progressions and consequences of pain. Quadriceps function and self-reported function were inversely related to both severity and duration of AKP (Table 6), which is also in line with earlier work.^16,18,19,25  Although self-reported function could be impaired without muscle dysfunction (Table 5), the weak-to-moderate relationships (0.36 < r < 0.51) between these factors suggest that quadriceps-strengthening exercises²⁷  could be a key intervention when treating patients with AKP. Similar to our outcomes, strong negative relationships have been described between pain severity or duration and quadriceps activation (ie, Hoffmann reflex normalized by the motor response).¹⁹  An abnormal monosynaptic stretch reflex in the spinal cord may provide useful insight into spinally mediated involuntary mechanisms as a result of pain^10,11 ; however, the results of this stretch-reflex assessment—a laboratory-based technique^11,14,19  (ie, lying supine without voluntary contraction)—are difficult to apply in clinical practice or functional movements such as daily activities.¹⁴  In addition to these data,¹⁹  quadriceps function measures involving force-based techniques¹⁴  (ie, maximal [MVIC] and explosive [RTD] strength, activation [CAR], and endurance [APT]) via voluntary contractions (ie, isometric and isokinetic) would offer clinicians further insight into which muscle capacity needs to be evaluated and treated before and after strengthening exercises are performed.

We analyzed a relatively homogeneous sample of participants by matching age, height, mass, body mass index, thigh circumference, and physical activity (Tables 2–5). Nevertheless, some limitations should be considered when interpreting our results. First, earlier authors⁸  identified that patients with AKP were less physically active (eg, fewer steps per day and less duration) compared with their healthy counterparts. Therefore, it is difficult to generalize our results to patients with AKP who are inactive (eg, exercise <150 min/wk). Nonetheless, our patients were as physically active as the control healthy population, which allowed objective comparison. Second, our testing procedures with the SIB for quadriceps activation could have induced or increased pain perception. This type of pain directly corresponds with the outcomes, which we did not obtain. Future investigators should examine how usual pain or pain during testing is associated with the quadriceps function. Third, as in other chronic AKP studies,^19,22  recall bias as to the exact date of the atraumatic onset of pain may be a factor. Fourth, the self-reported function was measured using the LEFS, which is not specific to AKP. This was intentional to assess an overall function of the lower extremity. Additionally, the LEFS has high measurement reliability and moderate responsiveness in patients with AKP.²⁹  Last, because of our study design, we were unable to track and record how pain severity changed over time (ie, pain variability) and how many asymptomatic periods existed between the onset of pain and the week before testing. As the authors³⁸  of a recent study identified that pain variability may predict subjective function in this population, we acknowledge that this may have influenced our findings. This aspect could be examined in prospective cohort studies to understand when and how neuromuscular function and self-reported function change with time-dependent pain variability.

Both the severity and duration of AKP were inversely associated with quadriceps function and self-reported function. Furthermore, higher perceived (eg, >3 of 10) and longer-lasting (eg, >2 years) AKP, especially when combined, may cause additional deficits in quadriceps function. Clinicians should consider both pain severity and duration when rehabilitating patients with AKP and be especially aware of the combined effect of these factors.