Various pathologies affecting the lower extremity, including plantar heel pain[1–3], metatarsalgia[4, 5], stress fractures of the foot, and Achilles tendionpathy[6] are associated with limited ankle dorsiflexion (ADF). During gait reduced ADF results in an increase of forefoot pressure, which might be responsible for the above outlined pathologies[7–9]. Studies were able show, that increasing ADF in these patients leads to a reduction of the symptoms[10–13]. ADF can be impaired due to osseous, ligmentous, neurologic or muscular restrains, with musculus gastrocnemius tightness (MGT) being the most common cause[1, 14].

The musculus gastrocnemius has an influence on ADF because it bridges the knee and ankle joint. Under physiological conditions the gastrocnemius is under full tension when the knee is extended, as the muscle’s origin is furthest from its insertion. ADF is then restrained by the muscle’s tension. On the contrary, knee flexion increases ADF, as the muscle’s origin and insertion are approximated. Further ADF is then limited by other structures of the ankle joint (Figure 1). In symptomatic patients the first assessment should therefore evaluate ADF with the knee fully extended. In case of impaired ADF one should then identify whether ADF can be increased by knee flexion. Patients with MGT demonstrate a reduced ADF with the knee fully extended, but ADF can be increased by knee flexion[2]. Identification of isolated MGT is essential for both physiotherapists and physicians, as it can be treated by stretching or endoscopic musculus gastrocnemius recession. If ADF cannot be increased by knee flexion the gastrocnemius is not responsible for the impaired ADF.

Consequently, any clinical test investigating MGT compares ADF with the knee extended to ADF with the knee flexed. This procedure was first described in 1923 by Nils Otto Silfverskiöld, an Swedish orthopedic surgeon[15].

In the literature, numerous tests are described to assess ADF. These tests can be categorized into non weightbearing[3, 16], weightbearing[17, 18], and instrumented[19, 20]. In everyday practice non weightbearing measurements are most commonly performed[21]. Nevertheless, evidence suggests a higher intra- and interrater reliability for weightbearing compared to non weightbearing measurements[17, 18, 21–23]. Furthermore, maximum ADF significantly differs from weightbearing to non weightbearing[23].

Independent of the test applied, no standard has been defined for the degree of knee flexion needed to eliminate the effect of the musculus gastrocnemius on ADF. Most studies conducting non weightbearing measurements applied a knee flexion of 90°[3, 7, 16]. On the contrary, most weightbearing measurements do not control for knee flexion[16, 23]. Furthermore, weightbearing measurements with 90° knee flexion are not feasible. Consequently, it is of upmost importance to identify the minimal required degree of knee flexion to eliminate the effect of the musculus gastrocnemius on ADF.

To our best knowledge the degree of knee flexion needed to eliminate the restraining effect of the gastrocnemius on ADF is unknown. Therefore, the aim of this study was to identify the influence of varying amounts of knee joint flexion on ADF.

Musculus gastrocnemius tightness (MGT) is the most common cause for impaired dorsiflexion of the ankle joint[1, 14]. It is held responsible for various pathologies affecting the lower extremity[4–6]. Identification of this cause by the treating physiotherapist or physician is extremely important as it can be easily treated by gastrocnemius lengthening either through physiotherapy or surgery. Various tests try to assess MGT by comparing ADF with the knee extended and flexed[3, 18, 20]. None of these tests has identified the degree of knee flexion needed to eliminate the effect of the gastrocnemius on ADF. Our study is the first to investigate the influence of the knee’s position on ADF. By 20° of knee flexion, the ADF restraining effect of a tight gastrocnemius on was already eliminated such that further knee flexion provided no additional ADF.

In the literature, different anatomical landmarks are described to assess the range of motion of the ankle joint. The landmarks most frequently used are the long axis of the fibula and either the plantar surface of the foot[21, 31–33] or the axis of the fifth metatarsal bone[16, 24–27, 34–36]. During pretests, we found the accurate identification of the plantar aspect of the foot to be highly dependent on the hand position of the examiner. Therefore, we chose the long axis of the fibula and the fifth metatarsal bone as references for non weightbearing measurements. For weightbearing tests, the floor clearly defined the x-axis, but the long axis of the fibula was hard to be identified due to the prominence of the peroneal tendons while standing. We therefore decided to mark the long axis of the fibula and the fifth metatarsal bone prior to testing[24, 28]. This procedure was also employed by Astroem and Arvidson[28] who conducted range of motion measurements to the foot in 121 healthy subjects. One has to keep in mind, that this might have a positive impact on the interrater reliability. In our study the interrater ICC values ranged from 0.961 to 0.992. These values are excellent, especially when compared to other non weightbearing measurements, with values ranging between 0.29[36] and 0.81[37]. Comparably good values have only been reported for weightbearing ADF measurements[17, 38]. Our goal was to clearly define the degree of knee flexion needed to eliminate the effect of the musculus gastrocnemius on ankle dorsiflexion. Consequently, it was crucial to use a reliable measurement technique. The herein observed interclass correlation coefficient argues for our standardized measurement protocol.

We observed a great variation of ADF between the subjects with values ranging for the extended knee from -11° to 23° non weightbearing and from 16° to 40° weightbearing. For the knee 75° flexed values ranged from 9° to 49° non weightbearing and from 31° to 58° weightbearing. Comparable variations for ADF have been reported in the literature. Non weightbearing values range from -2° ± 5°[16] to 20° ± 5°[3] for the knee extended and from 12° ± 6°[16] to 25° ± 5°[3] for the knee flexed. Furthermore, we could observe significant differences between weightbearing and non weightbearing ADF measurements. This goes well in line with the observation of other authors[2, 16, 18, 21, 23, 28]. Although there is less data available for weightbearing measurements, reported values range from 21° ± 7°[21] to 39° ± 5°[18] for the knee extended and from 33° ± 7°[16] to 50° ± 6°[23] for the knee flexed. The great range of ADF values might be due to the heterogeneity of the measurement procedures and the different landmarks used. Another pitfall we observed during the pretests was that minimal knee flexion had a profound impact on the ADF when measuring the ADF with the knee extended. Consequently, for our testing procedure one observer ensured full extension of the knee during measurements. Moreover, the great range of ADF might simply reflect the constitutional variation within the population.

Limitations that should be discussed are adjacent joint movements affecting ADF and the measurement device used. ADF is not limited to the tibiotalar joint, but also occurs partially in the subtalar and midtarsal joints. We tried to account for this problem by maintaining the foot in a subtalar neutral position, as recommended by previous authors[39–42]. Second, the goniometer used has 2° increments. Although other devices might be more accurate, the goniometer has to be considered the clinical gold standard[30]. With respect to the high ICC observed we believe this tool to be sufficiently accurate. Furthermore, because 20 degrees was the smallest knee flexion angle tested after full knee extension, it might be possible that even less knee flexion is already sufficient.. Nevertheless, we believe that every patient is capable to perform a Lunge test with 20° of knee flexion.

As stated above decreased ADF is held responsible for a variety of disorders. Up to date, there is neither a consensus on the degree of ADF considered pathological, nor whether tests should be conducted non weightbearing or weightbearing. The latter though has a pronounce impact on the degree of ADF measured. Weightbearing measurements have several advantages. First, they can be conducted by a single investigator. Second, they are independent of the torque applied by the observer, being a possible source of bias, and in addition more closely reflect the physiological torque during gait. Third, they have been shown to be more reliable[17, 18, 21–23]. Considering these aspects weightbearing tests should become the clinical standard. Nevertheless, most patients cannot conduct weightbearing tests with the knee 90° flexed, as conducted in non weightbearing measurements[3, 7, 16]. Consequently, it is of upmost importance to identify the minimal degree of knee flexion needed to eliminate the ADF restraining effect of the musculus gastrocnemius. This study is the first to investigate this problem. We were able to clearly demonstrate that already 20° of knee flexion sufficiently eliminates the effect of the musculus gastrocnemius on ADF.

When assessing MGT the degree of knee flexion needed to eliminate the restraining effect of the musculus gastrocnemius on ADF was unknown. We were able to demonstrate that already 20° of knee flexion fully eliminates the restraining effect of the musculus gastrocnemius on ADF, both non weightbearing and weightbearing. Our results build the bases to define a standardized clinical examination for musculus gastrocnemius tightness.