Skip to main content
Journal of Experimental Orthopaedics
Download PDF
  • Original Paper
  • Open access
  • Published:

Near-Infrared Spectroscopy measurements are reliable for studying patellar bone hemodynamics and affected by venous occlusion, but not by skin compression

Journal of Experimental Orthopaedics volume 10, Article number: 124 (2023) Cite this article

Abstract

Purpose

According to the homeostasis model, patellofemoral pain (PFP) results from disturbed homeostasis due to vascular insufficiency in the anterior knee. Near-Infrared Spectroscopy (NIRS) measures relative changes in concentrations (in µmol/cm2) of (de-)oxygenated hemoglobine (HHb and O2Hb). The aims were to: 1) investigate the characteristics of the NIRS signal derived from the patella during experiments affecting hemodynamics in healthy controls, and 2) determine the test–retest reliability of NIRS in positions clinically relevant for PFP patients.

Methods

Two experiments were conducted on 10 healthy controls and analysed using Student’s t-test. Reliability (ICC2,1) was evaluated for two activities (‘Prolonged Sitting’ and ‘Stair Descent’) in five PFP patients and 15 healthy controls, performed twice within five days.

Results

The NIRS signal (HHb and O2Hb) showed a statistically significant increase (p < .001 – .002) on all optodes (30, 35, 40 mm) during ‘Venous Occlusion’ (M = 1.0 – 2.0), while it showed no statistically significant change (p = .075 – .61) during ‘Skin Compression’ (M = -0.9 – 0.9) on the 30 and 35 mm optode. Reliability of NIRS (HHb and O2Hb) ranged from moderate to almost perfect (ICC2,1 = .47 – .95) on the 30 mm optode for ‘Prolonged Sitting’, and from moderate to substantial (ICC2,1 = .50 – .68) on the 35 mm optode for ‘Stair Descent’.

Conclusions

Patella NIRS measurements are affected by venous occlusion, but not by skin compression, and are sufficiently reliable as research application to compare real-time patellar bone hemodynamics. These insights may assist to improve effectiveness of evidence-based treatment strategies for PFP.

Trial registration

ISRCTN Trial Registration under number: 90377123.

Background

Patellofemoral pain (PFP) is a common and difficult to treat clinical condition. The annual prevalence in the general population ranges from 16 to 29% [48]. Despite evidence-based treatment recommendations advocating multimodal interventions, including exercise therapy, more than 40% of PFP patients reportedly continue experiencing knee symptoms five to eight years after diagnosis [8, 24, 27, 29, 30].

Since the 1980s, patellofemoral maltracking (PFM) has been the most widely accepted model explaining the genesis of PFP [39]. This model refers to altered patellofemoral kinematics and contact forces, as it may be the result of impaired quadriceps function or lateral soft tissue tightness [29, 45]. While some studies have expressed serious methodological concerns about the validity of the PFM model [20, 27], the aforementioned multimodal treatment recommendations are all based on this model. This dependence might account for the poor long-term treatment outcomes.

According to the more recent homeostasis model [16, 39], PFP is the result of a disturbed homeostasis of structures of the anterior part of the knee, including soft and osseous tissues, due to supraphysiologic loading [39]. Although no data from larger prospective trials conclusively permit causal inferences on the validity of this model, evidence aligning with the homeostasis model consists of morphologic changes of the retinacula [26, 45], including neovascularisation and hyperinnervation [40, 41], and increased intraosseous water content and pressure [22, 33]. Some authors suggest that these changes result from hypoxia due to vascular insufficiency of the peripatellar anastomotic ring [19, 40, 42, 47]. Only a few cross-sectional studies have examined patellar blood flow in PFP patients, suggesting reduced drainage time of the venous system, reduced pulsatile blood flow, and statistically non-significant differences in blood perfusion measured with intraosseous phlebography, photoplethysmography, and dynamic contrast-enhanced magnetic resonance imaging (MRI), respectively [2, 36, 51].

Near-Infrared Spectroscopy (NIRS) is an optical, non-invasive method using a light-source and -detector that enables continuous assessment of bone hemodynamics [5, 32]. While NIRS is an established tool in muscle physiology and transcranial cerebral circulation [43, 52], it has not been used to evaluate homeostasis of the patella in PFP patients.

Before studying differences in hemodynamics between PFP patients and healthy controls, the current study aimed to: 1) investigate the characteristics of the NIRS signal derived from the patella during interventions affecting hemodynamics of the anterior knee in healthy controls, and 2) determine the test–retest reliability of NIRS derived from the patella in positions clinically relevant for PFP patients.

Methods

Participants

This observational study (level of evidence III) complied with the requirements of the Declaration of Helsinki [54]. The study protocol obtained approval by the Medical Research Ethics Committee of Amsterdam UMC location University of Amsterdam under number NL77408.018.21 and has been registered at the ISRCTN registry under number 90377123 prior to the start of the data collection.

A convenient sample was obtained by recruiting subjects from 1) a private physical therapy clinic, 2) the physical therapy education program of the HAN University of Applied Sciences and 3) the Radboud University in Nijmegen, the Netherlands. Recruitment took place in February and March 2022. A senior physical therapist (MO) screened subjects for eligibility through history taking and standardised physical examination. Criteria for inclusion and exclusion are detailed in Table 1.

Table 1 Criteria for Inclusion and Exclusion
Full size table

After inclusion, participants´ demographic data including gender, age, body mass index (BMI), current smoking status, and hours of sports per week were collected. Blood pressure was measured using an automatic blood pressure monitor (Omron M6; Omron Healthcare, Kyoto, JP) [1]. Additionally, all participants completed the Tegner score and the Anterior Knee Pain Scale (AKPS) [25, 49]. The Tegner score assesses the current physical activity level on a scale from 0 to 10 [9, 23], with higher scores indicating higher activity levels. The Dutch version of the Tegner score is reliable (ICC = 0.93 – 0.97) and valid with an internal consistency of r = 0.73 – 0.83 [17]. The AKPS, a 13-item questionnaire, measures subjective symptoms and functional disabilities associated with PFP. Higher scores correspond to fewer symptoms and disabilities [11], and the Dutch version is reliable (ICC = 0.98) and valid with an internal consistency of r = 0.78 – 0.80 [50]. Participants were instructed not to participate in any sports activity 12 h before the NIRS measurement.

Near-Infrared Spectroscopy (NIRS)

Near-infrared light penetrates human tissue superficially with reported penetration depths up to 4 cm [4]. The NIRS device used was the PortaLite (Artinis Medical Systems; Elst, The Netherlands). The PortaLite is a continuous-wave device and the sensor, with a surface of 13.4 cm2, consists of three light emitting diodes arranged at 30, 35 and 40 mm distances from the detector. The diodes transmit near-infrared light at wavelengths of 760 and 820 nm, which have specific absorption characteristics for deoxygenated (HHb) and oxygenated hemoglobine (O2Hb). Given the measurement depth of NIRS (approximately half the sensor-detector distance [3]), this device measures HHb and O2Hb concentrations between 15 and 20 mm tissue depth. Superficial tissue (e.g., skin) has a non-significant confounding contribution on NIRS measurements if the superficial layer is less than 4 mm [18].

Evaluation of HHb and O2Hb concentrations using NIRS relies on the Beer-Lambert law modified with the ‘differential pathlength factor’ (DPF) [3, 15], which considers measurements taken in biological tissue. As no research with NIRS was conducted to examine patellar hemodynamics, the DPF was estimated using the equation: \(\frac{1}{2}{\left(\frac{3{{\mu }{\prime}}_{s}\left(\lambda \right)}{{\mu }_{a}\left(\lambda \right)}\right)}^{1/2},\) which takes the absorption (μa) and reduced scattering (μs) coefficients of the human skull into account [44], resulting in a DPF of 8.66.

Data were sampled at a frequency of 10 Hz. NIRS measures relative changes of HHb and O2Hb concentrations, not absolute concentrations, compared to a baseline. Therefore, all measurements involved a 3-min baseline, and relative changes in HHb and O2Hb between baseline and experimental measurements (Δ) were calculated in µmol/cm2 [3]. The laboratory room maintained constant light, with a room temperature between 21 and 23℃.

Patella skinfold and width

For detailed anthropometric characteristics of the anterior knee, the prepatellar skinfold thickness and the patella width were measured. Marks were made with a skin marker (Edding 8020; Edding GmbH, Ahrensburg, GER) on the medial and lateral side of the widest part of the patella and a cross midway between these two marks. Additionally, the proximal edge of the patella was marked to avoid placing the sensor too proximal (Fig. 1, left knee), following guidelines from the International Society for the Advancement of Kinanthropometry (ISAK) [12]. After training by a certified and ISAK (level 1) registered dietitian, the skinfold thickness was measured (AW) at the cross at the center of the patella with a skinfold caliper (Harpenden; Baty International, West Sussex, UK) and the patella width at the widest part with a slide caliper (Innovare; Cescorf, Porto Alegre, BR) (Fig. 2a and b).

Fig. 1
figure 1

Marks on the skin of the left knee and sensor placement on the right knee

Full size image
Fig. 2
figure 2

a Skinfold measurement and 2b Patella width measurement

Full size image

Both measurements were conducted twice, with a third measurement obtained if the difference between the repeated measurements exceeded 5%. The arithmetic mean was calculated for all measurements. Subsequently, the NIRS sensor was affixed to the skin using medical transparent double-sided adhesive tape (type 2181; 3 M, St. Paul, MN, US) with its middle on the cross, the three light sources positioned laterally and the detector medially (Fig. 1, right knee). An opaque cloth covered the NIRS sensor to minimise ambient light influence on the signal.

Experiments

To investigate the characteristics of the NIRS signal derived from the patella, two experiments affecting hemodynamics of the anterior knee were conducted on the dominant leg of 10 healthy controls.

Experiment 1 ‘Venous Occlusion’

As disturbed venous outflow of the patella is a key feature of the homeostasis model [2], an experiment assessed changes of the NIRS signal under venous occlusion of the dominant leg. Using a thigh cuff (Leg Cuff V3.1; Fitcuffs, Odder, Denmark) positioned 10 cm proximal to the mark at the patella’s proximal edge, venous occlusion was achieved. Before the start of the experiment, participants lay in supine position for 15 min with the knee flexed in 20 degrees. Examiner AW set this angle with an extendable goniometer (Lafayette, IN, US). Examiner AW visually inspected the NIRS signal for stability and the 3-min baseline was started once stability was confirmed. Subsequently, the cuff was inflated to 60 mmHG for one minute, inducing venous occlusion without disturbing arterial blood flow (Fig. 3a) [21, 35].

Fig. 3
figure 3

a Experiment 1 ‘Venous Occlusion’ and 3b Experiment 2 ‘Skin Compression’. (Note: Absence of the opaque cloth to visualise the set-up. During measurements the sensor was covered)

Full size image

Experiment 2 ‘Skin Compression’

Based on the previously described light-detector distance of the NIRS optode, the patellar bone is the main target tissue of the NIRS measurement. By creating skin compression, blood flow of the skin is compromised, but the NIRS signal is expected not to change significantly. Skin compression was achieved with a handheld dynamometer (Biofet; Mustec BV, Almere, The Netherlands). When the NIRS signal was stable (visual inspection), the baseline of Experiment 2 ‘Skin Compression’ was started three minutes after finishing Experiment 1 ‘Venous Occlusion’. After the 3-min baseline the handheld dynamometer was placed on the NIRS sensor to create mechanical skin compression of 100 mmHG (135 gr/cm2 or 1.8 kg) for one minute (Fig. 3b) [35]. After finishing both experiments, NIRS measurements were continued for 15 s (recovery time).

Test–Retest reliability

The test–retest reliability of NIRS in the patellar bone was evaluated for two clinically relevant activities for PFP patients: prolonged sitting with the knees flexed and stair descent. Despite the absence of active loading of the patellofemoral joint, more than 80% of PFP patients experience pain with prolonged sitting [10]. Previous research recommends to explore the mechanism of sitting pain [10]. Additionally, stair climbing is among the most provoking activities of daily living for PFP patients [6]. To evaluate the test–retest reliability five PFP patients and 15 healthy controls performed both clinically relevant activities twice with three to five days in between.

Activity 1 ‘Prolonged Sitting’

After a 15-min rest period with the participants’ knees flexed at 20 degrees and stable NIRS signal (visual inspection), a 3-min baseline was established in the same position. Subsequently, knees were flexed to 90 degrees for 30 min (Fig. 4a), followed by returning to 20 degrees of flexion for five minutes. Knee angles were verified using an extendable goniometer (Lafayette, IN, US).

Fig. 4
figure 4

a Activity 1 ‘Prolonged Sitting’ and 4b Activity 2 ‘Stair Descent´. (Note: Absence of the opaque cloth to visualise the set-up, during measurements the sensor was covered)

Full size image

Activity 2 ‘Stair Descent’

Stair descent was simulated using the previously developed Decline Step-Down Test (DSDT) [37]. Following a 15-min rest in a standing position and ensuring stable NIRS signal (visual inspection), a 3-min baseline was established while standing on the decline step-down set-up with extended knees. Participants performed a motion simulating stair descent (45 degrees of knee flexion) and held this for one minute (Fig. 4b). Correctness of this knee angle was confirmed with an extendable goniometer (Lafayette, IN, US). After a 3-min rest, the procedure was repeated for the other leg, with the starting leg randomised resulting in half of the participants starting with the left and the other half with the right leg.

Blinding

Blinding for health status (PFP patient or healthy control) was assured. Examiner MO was responsible for participant inclusion and exclusion, while examiner AW conducted measurements without knowledge of health status.

Sample size

Given the absence of information regarding expected effect sizes in NIRS studying patellar bone hemodynamics, the sample size was maximised based on available resources. A convenient sample of five PFP patients (5–10 symptomatic knees) and 15 healthy participants (30 knees) was anticipated.

Statistical analysis

Normality of distribution was assessed using visual inspection and the Shapiro–Wilk test. The participants’ demographic data were described using means (M), standard deviations (SD) for continuous variables, and percentages (%) for dichotomous variables. Differences in continuous baseline characteristics between PFP patients and healthy controls were analysed using Student’s t-test for normally distributed data or Mann–Whitney U-test for not normally distributed data. Differences in categorical baseline characteristics were analysed using Fisher’s exact test.

For both experiments, mean and standard deviation for changes in HHb and O2Hb between baseline and experimental measurement (ΔBas_Exp1 and ΔBas_Exp2) were calculated separately for each of the three optodes (30, 35, 40 mm). The first and last three seconds were removed to eliminate any influence of movement artifacts. Additionally, concentrations of HHb and O2Hb on the three optodes were visualised with plots.

Effect size (ES) was calculated using Cohen’s d to present the magnitude of the change between baseline and experimental measurement. An ES of 0.2 was considered small, 0.5 medium, and ≥ 0.8 large [7].

A NIRS sensor was placed on both knees of each participant. For the statistical analysis, the knees were included separately. For PFP patients, only the symptomatic knees were included, excluding non-symptomatic knees from the statistical analysis. Both knees of healthy controls were included.

Test–retest reliability of NIRS in the patellar bone was assessed for both activities. For Activity 1 ‘Prolonged Sitting’ four changes in HHb and O2Hb were assessed: 1) change between baseline and first 10 min of sitting with knees flexed (ΔBas_Sit0-10), 2) change between baseline and second 10 min of sitting with knees flexed (ΔBas_Sit10-20), 3) change between baseline and third 10 min of sitting with knees flexed (ΔBas_Sit20-30), 4) change between third 10 min of sitting with knees flexed and five minutes of recovery time (ΔSit20-30_Rec). For Activity 2 ‘Stair Descent’, the change in HHb and O2Hb between baseline and 45 degrees of knee flexion was assessed (ΔBas_Step).

Changes were calculated for each concentration (HHb, O2Hb) and optode separately (30, 35, 40 mm). Test–retest reliability was assessed using intraclass correlation coefficient (ICC), standard error of the measurement (SEM), and the smallest detectable change (SDC) for each change in HHb and O2Hb. Moreover, the ICC, SEM and SDC were determined for PFP patients and healthy controls separately. ICC2,1 was calculated using a single-rating, absolute agreement, two-way mixed effect model, with values indicating slight agreement (0.0 to 0.2), fair agreement (0.21 to 0.40), moderate agreement (0.41 to 0.60), substantial agreement (0.61 to 0.80), and almost perfect agreement (0.81 to 1.0) [28]. Measurement error as an indicator of the measurement's precision was assessed by the standard error of measurement (SEM) and calculated according to the formula: SD x√(1-ICC), SD being the standard deviation from mean scores and ICC for test–retest reliability [14]. The smallest detectable change (SDC) was calculated at group level, according to the formula: SEM × 1.96√2/√n [53].

NIRS data collection was performed with Oxysoft, version 3.2.72 (Artinis, Elst, The Netherlands). The Oxysoft files were converted into MATLAB files (R2020a version 9.8.0.1323502, Mathworks, Natick, USA). MATLAB was used to filter the raw values with a moving average filter, and calculate changes between baseline and experimental measurements. The outcomes were entered in Excel (Microsoft Office version 16), and the statistical analyses were performed using SPSS version 28.0 with a level of significance set at p < 0.05.

Results

Out of 21 subjects screened for participation, 20 were included, and one was excluded due to personal time constraints (Fig. 5).

Fig. 5
figure 5

Flowchart of the inclusion process

Full size image

Table 2 describes baseline characteristics of all participants. Fifteen healthy controls (10 females [66.7%], mean age 21.0 years [SD = 2.6] and mean BMI 22.0 kg/m2 [SD = 2.7]) and five patients with PFP (4 females [80.0%], mean age 22.2 years [SD = 2.6] and mean BMI 23.8 kg/m2 [SD = 3.1]) were included. Patients with PFP differed only in the AKPS score compared to healthy controls. The mean AKPS score of PFP patients (M = 87.2; SD = 3.8) was significantly lower compared to healthy controls (M = 98.3; SD = 2.1), (t(18) = 8.81; p < 0.001, d = 3.62), indicating correct group allocation.

Table 2 Baseline characteristics
Full size table

Experiments

Both experiments were conducted on 10 consecutively enrolled healthy participants before their performance of Activity 1 ‘Prolonged Sitting’ and Activity 2 ‘Stair Descent’. All participants completed both activities during the two visits as planned. Table 3 presents the mean and standard deviation for the change in HHb and O2Hb between baseline and experimental measurements. Significant increases in both HHb and O2Hb concentrations were observed on all optodes (30, 35, 40 mm) during Experiment 1 ‘Venous Occlusion’ (p < 0.001 – 0.002). Additionally, statistically significant increases in HHb and O2Hb were noted only on the 40 mm optode during Experiment 2 ‘Skin Compression’ (p = 0.004 – 0.006). No statistically significant changes (p = 0.075 – 0.61) were observed on the 30 mm and 35 mm optodes during Experiment 2 ‘Skin Compression’. Plots of HHb and O2Hb during both experiments are presented in Fig. 6.

Table 3 Changes in HHb and O2Hb for Experiment 1 ‘Venous Occlusion’ and Experiment 2 ‘Skin Compression’
Full size table
Fig. 6
figure 6

Plotted means of HHb and O2Hb (in µmol/cm.2) of Experiment 1 ‘Venous Occlusion’ and Experiment 2 ‘Skin Compression’. Abbreviations: HHb, deoxygenated hemoglobine; O2Hb, oxygenated hemoglobine for each optode (30, 35, 40 mm); min, minutes; A, baseline (3 min); B, experimental measurement (venous occlusion / skin compression) (1 min); C, recovery time (15 s)

Full size image

Test–retest reliability of Activity 1 ‘Prolonged Sitting’

Table 4 shows results of reliability analysis for HHb for each optode (30, 35, 40 mm). Overall, ICCs for the 30 mm optode indicated moderate to substantial agreement for total (n = 39), healthy control (n = 30) and patient sample (n = 9). However, ICCs for the 35 and 40 mm optode showed moderate to substantial agreement for total and healthy control sample but slight to fair agreement in the patient sample.

Table 4 Test–retest reliability of Activity 1 ‘Prolonged Sitting’ (HHb)
Full size table

For O2Hb (Table 5), ICCs for the 30 mm and 40 mm optodes indicated moderate to almost perfect agreement for total, healthy control and patient sample. However, ICCs for the 35 mm optode showed moderate to substantial agreement for total and patient sample but fair agreement in the healthy control sample for the change between baseline and second 10 min of sitting with knees flexed.

Table 5 Test–retest reliability of Activity 1 ‘Prolonged Sitting’ (O2Hb)
Full size table

Test–retest reliability of Activity 2 ‘Stair Descent’

Table 6 presents results of reliability analysis for HHb for each optode (30, 35, 40 mm). ICCs for the 35 mm optode showed moderate to substantial agreement for total, healthy control and patient sample. However, ICCs for the 30 mm and 40 mm optode showed substantial agreement for total and healthy control sample but slight to fair agreement in the patient sample.

Table 6 Test–retest reliability of Activity 2 ‘Stair Descent’ (HHb)
Full size table

For O2Hb (Table 7), ICCs for the 35 mm optode showed moderate to substantial agreement for total, healthy control and patient sample. However, ICCs for the 30 mm and 40 mm optode showed moderate agreement for total and healthy control sample but slight to fair agreement in the patient sample.

Table 7 Test–retest reliability of Activity 2 ‘Stair Descent’ (O2Hb)
Full size table

Discussion

This study, for the first time, investigated characteristics of the NIRS signal during experiments affecting hemodynamics of the patella in healthy controls and to evaluate test–retest-reliability in both PFP patients and healthy controls. The main findings revealed that 1) venous occlusion significantly increased the NIRS signal (HHb and O2Hb) derived from the patella, while no significant changes occurred during skin compression, and 2) the test–retest reliability of NIRS measures (HHb and O2Hb) derived from the patella showed moderate to almost perfect agreement in positions clinically relevant for PFP patients. These findings indicate that while NIRS measurements of the patella are sufficiently reliable as research application to compare real-time bone hemodynamics in PFP patients and healthy controls, they might not be suitable for clinical application for individual subject evaluation.

Venous occlusion of the thigh resulted in an increase of the NIRS signal (HHb and O2Hb) on all optodes (30, 35, 40 mm). With venous occlusion, there is (arterial) inflow while (venous) outflow is compromised, which results in increased concentrations of HHb and O2Hb. This is in line with previous measurements with this device [3] and other studies assessing blood flow in the forearm with NIRS and strain-gauge plethysmography [13, 34]. Näslund et al. [35] found reduced intraosseous blood flow of the patella using photoplethysmography during venous occlusion. Experiment 1 ‘Venous Occlusion’ mimicked vascular insufficiency and observed NIRS signal changes could be expected when comparing bone hemodynamics of the patella in PFP patients and healthy controls.

The NIRS signal (HHb and O2Hb) did not increase on the 30 mm and 35 mm optode when skin compression of the patella was applied. This is in line with our hypothesis, since the measurement depth of this NIRS device (15 mm to 20 mm) is deeper than the thickness of the skin. The observed mean prepatellar skinfold thickness was 7.7 mm and 7.1 mm in PFP patients and healthy controls, resulting in skin thickness (skinfold dividing by two) of 3.9 mm and 3.6 mm, respectively. A skin thickness of less than 4 mm has a non-significant confounding contribution on the NIRS signal [18]. Surprisingly, a statistically significant (p = 0.004 – 0.006) increase of the NIRS signal was found on the 40 mm optode. More research is needed on this specific topic.

Furthermore, the patella width in the current study was between 50.1 mm (patient sample) and 51.4 (total sample). Porteus et al. [38] reported a ‘Normal Index of Patella Width: Thickness’ of 1.8:1, which translates into an estimated patella thickness of 27.8 mm to 28.6 mm including cartilage. We conclude that even the 40 mm optode (measurement depth of 20 mm) still measures in the patellar bone.

The ICCs of the 30 mm optode (0.47 – 0.95) showed moderate to almost perfect agreement during Activity 1 ‘Prolonged Sitting’ and therefore this optode is most suitable to compare changes of HHb and O2Hb concentrations in PFP patients and healthy controls. The ICCs of the 35 mm optode (0.50 – 0.68) showed moderate to substantial agreement during Activity 2 ‘Stair Descent’ and therefore this is most suitable to compare changes of HHb and O2Hb concentrations in PFP patients and healthy controls. Notably, in PFP patients during Activity 1 ‘Prolonged Sitting” the ICCs for HHb of the 35 mm and 40 mm optode indicated slight to fair agreement only, while the ICCs for O2Hb indicated almost perfect agreement on all optodes. Given the large 95% confidence intervals this observed difference may be due to the small sample size (n = 9). This also applies to the fair reliability of the 30 mm and 40 mm optodes found during Activity 2 ‘Stair Descent’ in PFP patients.

Generally, this study found wide variability of measures of reliability for both Activity 1 ‘Prolonged Sitting’ and Activity 2 ‘Stair Descent’. Based on results of the current study, it is unclear what is attributable to normal biological variation of hemoglobine concentrations in bone tissue or to measurement error. Regarding biological variability, heterogeneity of bone tissue and lower metabolic rate than muscle tissue, including lower sampling volume of HHb and O2Hb in bone, are known barriers in vascular measurement of bone [32].

Regarding measurement error, a random source of bias could be introduced by the determination of degrees of knee flexion during NIRS measurements by goniometer assessments. Measurement errors up to 10 degrees have been reported [31]. Generally, there are no known optode-specific sources of bias and all optodes measure changes of HHb and O2Hb concentrations at a tissue depth of 15—20 mm [3]. All other known sources of bias like no exercising 24 h before and rest time prior to measurements, and standardisation of optode placement have been taken into account. Due to this observed wide variability of measures of reliability, NIRS cannot play a role in the evaluation of patellar blood flow in the clinical setting on an individual level, but is limited to the cautious application in scientific research to evaluate differences in patellar blood flow on a group level.

Other clinical studies on blood flow of the anterior knee in PFP patients examined skin temperature by thermal imaging (only superficially) [46], pulsatile blood flow by photoplethysmography (only for five minutes) [36], or blood perfusion by dynamic contrast-enhanced MRI (only in knee extension) [51]. Our study is the first to study real-time bone hemodynamics of the patella continuously and in clinically relevant positions for PFP patients. We suggest that this protocol also can be used to study hemodynamics of the patella in patients with other anterior knee pain conditions, e.g., patellofemoral osteoarthritis. Whether the clinimetric properties of the assessment are similar should be subject of further study.

The current study is not without limitations. As mentioned before, the number of test–retest observations of the patient sample is considered to be low compared to the number of observations of the healthy control sample. The low number of patient observations affects statistical power and increases confidence intervals of the ICCs. Unfortunately, even if effect sizes would have been known prior to the start of the current study, given the limited resources of our study group, it would not have been possible to increase the number of participating PFP patients. It is recommended to collect additional data on the reproducibility of this NIRS protocol in PFP patients.

Another limitation was the absence of randomisation of the order of the experiments. During preparation of the study, the NIRS signal required 10 min to normalise after skin compression while the same signal normalised within one minute after venous occlusion. To avoid unnecessary burden for the participants, we therefore decided to perform the experiments without randomisation.

The performance of Activity 2 `Stair Descent` mimics stair descent in the real world only in a limited way, since stair descent is much more dynamic, but also much more random, resulting in movement variability. Unfortunately, movement variability would introduce major artefacts into the NIRS measurement. Nevertheless, the chosen performance here creates patellofemoral joint loading in combination with active muscle contraction.

Conclusions

The current study found NIRS measurements of the patella to be affected by venous occlusion of the thigh, but not by skin compression. Furthermore, NIRS measurements seem sufficiently reliable as research application to compare real-time patellar bone hemodynamics in PFP patients and healthy controls. This opens a window of opportunity to determine differences in hemodynamics of the patella between patients and healthy controls and may help to improve effectiveness of evidence-based treatment strategies for PFP patients.

Availability of data and materials

Research data will be available upon reasonable request.

Abbreviations

AKPS :

Anterior Knee Pain Scale

BMI:

Body Mass Index

CI :

Confidence Interval

DPF:

Differential Pathlength Factor

DSDT :

Decline Step-Down Test

ES :

Effect Size

HHb :

Deoxygenated Hemoglobine

ICC :

Intraclass Correlation Coefficient

ISAK :

International Society for the Advancement of Kinanthropometry

LLROM :

Lower Limb Range Of Motion

NIRS :

Near-Infrared Spectroscopy

O2Hb :

Oxygenated Hemoglobine

PFM :

Patellofemoral Maltracking

PFP :

Patellofemoral Pain

SD :

Standard Deviation

SDC :

Smallest Detectable Change

SEM :

Standard Error of the Measurement

VAS :

Visual Analogue Scale

References

  1. Altunkan Ş, Iliman N, Kayatürk N, Altunkan E (2007) Validation of the Omron M6 (HEM-7001-E) upper-arm blood pressure measuring device according to the International Protocol in adults and obese adults. Blood Press Monit 12:219–225

    Article  PubMed  Google Scholar 

  2. Arnoldi CC (1991) Patellar pain. Acta Orthop Scand Suppl 244:1–29

    Article  CAS  PubMed  Google Scholar 

  3. Artinis Medical Systems (2016) Manual Portalite for Oxysoft 3.0.95 and higher - Version 1610

    Google Scholar 

  4. Aziz SM, Khambatta F, Vaithianathan T et al (2010) A near infrared instrument to monitor relative hemoglobin concentrations of human bone tissue in vitro and in vivo. Rev Sci Instrum 81:043111

    Article  PubMed  Google Scholar 

  5. Binzoni T, van de Ville D (2011) Noninvasive probing of the neurovascular system in human bone/bone marrow using near-infrared light. J Innov Opt Health Sci 04:183–189

    Article  Google Scholar 

  6. Brechter JH, Powers CM (2002) Patellofemoral joint stress during stair ascent and descent in persons with and without patellofemoral pain. Gait Posture 16:115–123

    Article  PubMed  Google Scholar 

  7. Cohen J (1988) Statistical power analysis for the behavioral sciences, 2nd edn. Lawrence Erlbaum Associates, Hillsdale, NJ

    Google Scholar 

  8. Collins NJ, Barton CJ, van Middelkoop M et al (2018) 2018 Consensus statement on exercise therapy and physical interventions (orthoses, taping and manual therapy) to treat patellofemoral pain: recommendations from the 5th International Patellofemoral Pain Research Retreat, Gold Coast, Australia, 2017. Br J Sports Med 52:1170–1178

    Article  PubMed  Google Scholar 

  9. Collins NJ, Misra D, Felson DT et al (2011) Measures of knee function: International Knee Documentation Committee (IKDC) Subjective Knee Evaluation Form, Knee Injury and Osteoarthritis Outcome Score (KOOS), Knee Injury and Osteoarthritis Outcome Score Physical Function Short Form (KOOS-PS) Knee Ou. Arthritis Care Res (Hoboken) 63(Suppl 1):S208–S228

    PubMed  Google Scholar 

  10. Collins NJ, Vicenzino B, van der Heijden RA, van Middelkoop M (2016) Pain During Prolonged Sitting Is a Common Problem in Persons With Patellofemoral Pain. J Orthop Sports Phys Ther 46:658–663

    Article  PubMed  Google Scholar 

  11. Crossley KM, Bennell KL, Cowan SM, Green S (2004) Analysis of outcome measures for persons with patellofemoral pain: which are reliable and valid? Arch Phys Med Rehabil 85:815–822

    Article  PubMed  Google Scholar 

  12. da Silva VS, Vieira MFS (2020) International Society for the Advancement of Kinanthropometry (ISAK) Global: international accreditation scheme of the competent anthropometrist. Revista Brasileira de Cineantropometria Desempenho Humano 22:1–6

    Google Scholar 

  13. de Blasi RA, Ferrari M, Natali A et al (1994) Noninvasive measurement of forearm blood flow and oxygen consumption by near-infrared spectroscopy. J Appl Physiol 76:1388–1393

    Article  PubMed  Google Scholar 

  14. de Vet HC, W; Terwee, Caroline B., Mokkink Lidwine B., Knol DL, (2017) Measurement in Medicine, 8th edn. Cambridge University Press, Cambridge

    Google Scholar 

  15. Delpy DT, Cope M, van der Zee P et al (1988) Estimation of optical pathlength through tissue from direct time of flight measurement. Phys Med Biol 33:1433

    Article  CAS  PubMed  Google Scholar 

  16. Dye SF (2005) The Pathophysiology of Patellofemoral Pain: A Tissue Homeostasis Perspective. Clin Orthop Relat Res 436:100–110

    Article  Google Scholar 

  17. Eshuis R, Lentjes GW, Tegner Y et al (2016) Dutch Translation and Cross-cultural Adaptation of the Lysholm Score and Tegner Activity Scale for Patients With Anterior Cruciate Ligament Injuries. J Orthop Sports Phys Ther 46:976–983

    Article  PubMed  Google Scholar 

  18. Franceschini MA, Fantini S, Paunescu LA et al (1998) Influence of a superficial layer in the quantitative spectroscopic study of strongly scattering media. Appl Opt 37:7447–7458

    Article  CAS  PubMed  Google Scholar 

  19. Fulkerson JP (1983) The etiology of patellofemoral pain in young, active patients: a prospective study. Clin Orthop Relat Res 179:129–133

    Article  Google Scholar 

  20. Grant C, Fick CN, Welsh J et al (2020) A Word of Caution for Future Studies in Patellofemoral Pain: A Systematic Review With Meta-analysis. Am J Sports Med 49:539–551

    Google Scholar 

  21. Groothuis JT, van Vliet L, Kooijman M, Hopman MTE (2003) Venous cuff pressures from 30 mmHg to diastolic pressure are recommended to measure arterial inflow by plethysmography. J Appl Physiol 95:342–347

    Article  PubMed  Google Scholar 

  22. Ho K-Y, Hu HH, Colletti PM, Powers CM (2014) Running-induced patellofemoral pain fluctuates with changes in patella water content. Eur J Sport Sci 14:628–634

    Article  PubMed  Google Scholar 

  23. Howe TE, Dawson LJ, Syme G et al (2012) Evaluation of outcome measures for use in clinical practice for adults with musculoskeletal conditions of the knee: A systematic review. Man Ther 17:100–118

    Article  PubMed  Google Scholar 

  24. Kastelein M, Luijsterburg PAJ, Heintjes EM et al (2015) The 6-year trajectory of non-traumatic knee symptoms (including patellofemoral pain) in adolescents and young adults in general practice: a study of clinical predictors. Br J Sports Med 49:400–405

    Article  CAS  PubMed  Google Scholar 

  25. Kujala UM, Jaakkola LH, Koskinen SK et al (1993) Scoring of patellofemoral disorders. Arthroscopy 9:159–163

    Article  CAS  PubMed  Google Scholar 

  26. Lack S, Anthony L, Noake J et al (2019) Medial and Lateral Patellofemoral Joint Retinaculum Thickness in People With Patellofemoral Pain: A Case-Control Study. J Ultrasound Med 38:1483–1490

    Article  PubMed  Google Scholar 

  27. Lack S, Neal B, de Oliveira SD, Barton C (2018) How to manage patellofemoral pain - Understanding the multifactorial nature and treatment options. Phys Ther Sport 32:155–166

    Article  PubMed  Google Scholar 

  28. Landis JR, Koch GG (1977) The measurement of observer agreement for categorical data. Biometrics 33:159–174

    Article  CAS  PubMed  Google Scholar 

  29. Lankhorst NE, Bierma-Zeinstra SMA, van Middelkoop M (2012) Risk factors for patellofemoral pain syndrome: a systematic review. J Orthop Sports Phys Ther 42:81–94

    Article  PubMed  Google Scholar 

  30. Lankhorst NE, van Middelkoop M, Crossley KM et al (2016) Factors that predict a poor outcome 5–8 years after the diagnosis of patellofemoral pain: a multicentre observational analysis. Br J Sports Med 50:881–886

    Article  CAS  PubMed  Google Scholar 

  31. McVeigh KH, Murray PM, Heckman MG et al (2016) Accuracy and Validity of Goniometer and Visual Assessments of Angular Joint Positions of the Hand and Wrist. J Hand Surg Am 41:e21–e35

    Article  PubMed  Google Scholar 

  32. Meertens R, Casanova F, Knapp KM et al (2018) Use of near-infrared systems for investigations of hemodynamics in human in vivo bone tissue: A systematic review. J Orthop Res 36:2595–2603

    Article  PubMed  Google Scholar 

  33. Miltner O, Siebert CH, Schneider U et al (2003) Patellar hypertension syndrome in adolescence: a three-year follow up. Arch Orthop Trauma Surg 123:455–459

    Article  CAS  PubMed  Google Scholar 

  34. Nagashima Y, Yada Y, Hattori M, Sakai A (2000) Development of a new instrument to measure oxygen saturation and total hemoglobin volume in local skin by near-infrared spectroscopy and its clinical application. Int J Biometeorol 44:11–19

    Article  CAS  PubMed  Google Scholar 

  35. Näslund J, Pettersson J, Lundeberg T et al (2006) Non-invasive continuous estimation of blood flow changes in human patellar bone. Med Biol Eng Comput 44:501–509

    Article  PubMed  Google Scholar 

  36. Näslund J, Walden M, Lindberg LG (2007) Decreased Pulsatile Blood Flow in the Patella in Patellofemoral Pain Syndrome. Am J Sports Med 35:1668–1673

    Article  PubMed  Google Scholar 

  37. Ophey MJ, Bosch K, Khalfallah FZ et al (2019) The decline step-down test measuring the maximum pain-free flexion angle: A reliable and valid performance test in patients with patellofemoral pain. Phys Ther Sport 36:43–50

    Article  PubMed  Google Scholar 

  38. Porteous A, Sullivan N, Murray J, Eldridge J (2013) How thick is the patella? A reproducible measure of patella width: thickness from adult MRI. Orthop Proc 95-B:46

    Google Scholar 

  39. Sanchis-Alfonso V, Ramírez-Fuentes C, Roselló-Sastre E, et al (2020) Pathophysiology of Anterior Knee Pain. In: Patellofemoral Pain, Instability, and Arthritis. Springer Berlin Heidelberg

  40. Sanchis-Alfonso V, Rosella-Sastre E (2000) Immunohistochemical Analysis for Neural Markers of the Lateral Retinaculum in Patients with Isolated Symptomatic Patellofemoral Malalignment. Am J Sports Med 28:725–731

    Article  CAS  PubMed  Google Scholar 

  41. Sanchis-Alfonso V, Roselló-Sastre E, Monteagudo-Castro C, Esquerdo J (1998) Quantitative Analysis of Nerve Changes in the Lateral Retinaculum in Patients with Isolated Symptomatic Patellofemoral Malalignment. Am J Sports Med 26:703–709

    Article  CAS  PubMed  Google Scholar 

  42. Sanchis-Alfonso V, Rosello-Sastre E, Revert F, Garcia A (2005) Histologic Retinacular Changes Associated With Ischemia in Painful Patellofemoral Malalignment. Orthopedics 28:593–599

    Article  PubMed  Google Scholar 

  43. Scheeren TWL, Schober P, Schwarte LA (2012) Monitoring tissue oxygenation by near infrared spectroscopy (NIRS): background and current applications. J Clin Monit Comput 26:279

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Scholkmann F, Wolf M (2013) General equation for the differential pathlength factor of the frontal human head depending on wavelength and age. J Biomed Opt 18:105004

    Article  PubMed  Google Scholar 

  45. Schoots EJM, Tak IJR, Veenstra BJ et al (2013) Ultrasound characteristics of the lateral retinaculum in 10 patients with patellofemoral pain syndrome compared to healthy controls. J Bodyw Mov Ther 17:523–529

    Article  PubMed  Google Scholar 

  46. Selfe J, Hardaker N, Thewlis D, Karki A (2006) An Accurate and Reliable Method of Thermal Data Analysis in Thermal Imaging of the Anterior Knee for Use in Cryotherapy Research. Arch Phys Med Rehabil 87:1630–1635

    Article  PubMed  Google Scholar 

  47. Selfe J, Harper L, Pedersen I et al (2003) Cold legs: a potential indicator of negative outcome in the rehabilitation of patients with patellofemoral pain syndrome. Knee 10:139–143

    Article  CAS  PubMed  Google Scholar 

  48. Smith BE, Selfe J, Thacker D et al (2018) Incidence and prevalence of patellofemoral pain: A systematic review and meta-analysis. PLoS ONE 13:e0190892

    Article  PubMed  PubMed Central  Google Scholar 

  49. Tegner Y, Lysholm J (1985) Rating systems in the evaluation of knee ligament injuries. Clin Orthop Relat Res 198:43–49

    Article  Google Scholar 

  50. Ummels PEJ, Lenssen AF, Barendrecht M, Beurskens AJHM (2017) Reliability of the Dutch translation of the Kujala Patellofemoral Score Questionnaire. Phys Res Int 22:e1649

    Google Scholar 

  51. van der Heijden RA, Poot DHJ, Ekinci M et al (2018) Blood perfusion of patellar bone measured by dynamic contrast-enhanced MRI in patients with patellofemoral pain: A case-control study. J Magn Reson Imaging 48:1344–1350

    Article  PubMed  PubMed Central  Google Scholar 

  52. Vardi M, Nini A (2008) Near-infrared Spectroscopy for Evaluation of Peripheral Vascular Disease. A Systematic Review of Literature. Eur J Vasc Endovasc Surg 35:68–74

    Article  CAS  PubMed  Google Scholar 

  53. Weir JP (2005) Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM. J Strength Cond Res 19:231–240

    PubMed  Google Scholar 

  54. World Medical Association (2013) World Medical Association Declaration of Helsinki: Ethical principles for medical research involving human subjects. JAMA 310:2191–2194

    Article  Google Scholar 

Download references

Acknowledgements

The authors thank all participants for their cooperation.

Patient consent statement

Written informed consent procedure according to the Medical Research Involving Human Subjects Act.

Funding

The authors received no financial support for conducting this research project.

Author information

Authors and Affiliations

  1. IJsveldFysio – Private Physical Therapy Clinic, Nijmegen, The Netherlands

    Martin J. Ophey

  2. Department of Orthopaedic Surgery and Sports Medicine, Amsterdam UMC Location University of Amsterdam, Meibergdreef 9, 1105 AZ, Amsterdam, The Netherlands

    Martin J. Ophey & Gino M. M. J. Kerkhoffs

  3. ESP Science and Education, Vienna, Austria

    Martin J. Ophey, Robert van den Berg & Igor J. R. Tak

  4. RU - Radboud University, Biomedical Sciences, Nijmegen, The Netherlands

    Anne Westerweel & Maxime van Oort

  5. FH Burgenland, Physical Therapy Department, University of Applied Science, Pinkafeld, Austria

    Robert van den Berg

  6. AIM - Austrian Institute of Management, Advanced Physiotherapy & Management, Eisenstadt, Austria

    Robert van den Berg

  7. Amsterdam Collaboration On Health & Safety in Sports (ACHSS), IOC Research Center, Amsterdam, The Netherlands

    Gino M. M. J. Kerkhoffs & Igor J. R. Tak

  8. Academic Center for Evidence-Based Sports Medicine (ACES), Amsterdam, The Netherlands

    Gino M. M. J. Kerkhoffs & Igor J. R. Tak

  9. Physiotherapy Utrecht Oost – Sports Rehabilitation and Manual Therapy, Utrecht, The Netherlands

    Igor J. R. Tak

Authors
  1. Martin J. Ophey
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Anne Westerweel
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Maxime van Oort
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Robert van den Berg
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Gino M. M. J. Kerkhoffs
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Igor J. R. Tak
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

MO, RvdB, IT and GK conceived the study. MO, AW, MvO, RvdB, IT and GK were responsible for the study design and analysis. MO, AW, MvO and RvB were responsible for data collection. All authors have made substantial revisions to earlier drafts and approved the final manuscript.

Corresponding author

Correspondence to Martin J. Ophey.

Ethics declarations

Ethics approval and consent to participate

By the Medical Research Ethics Committee of Amsterdam UMC location University of Amsterdam under number NL77408.018.21, The Netherlands.

Competing interests

None. The authors certify that they have no affiliations with or financial involvement in any organization or entity with a direct financial interest in the subject matter or materials discussed in the article.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ophey, M.J., Westerweel, A., van Oort, M. et al. Near-Infrared Spectroscopy measurements are reliable for studying patellar bone hemodynamics and affected by venous occlusion, but not by skin compression. J EXP ORTOP 10, 124 (2023). https://doi.org/10.1186/s40634-023-00709-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s40634-023-00709-6

Keywords