INJURY DIAGNOSIS TENDINOPATHY

only discrepancy in Gibbon et al’s methodology was the position used during the examination. Their patients were in a prone position with the ankle in full dorsiflexion whereas patients in the other studies were prone with the ankle in a neutral position, resting off the end of the bed. Full ankle dorsiflexion would disrupt neovascularisation and the measurement of the tendon thickness, but it would not affect the position of the thickening. Gibbon et al’s (41) findings are well supported by a large sample size of 118 patients, of whom a significant proportion (109 out of 118 tendons) showed thickness in the proximal two-thirds. The reasons for the varying observations are unknown.

Presence of neovascularisation Two studies by Öhberg and colleagues (28,42) and one by Alfredson et al (29) produced results showing that neovascularisation (Fig. 7) was always present in the Achilles tendinopathy cases they examined and always absent in normal tendons. All three studies were case–control studies with significant sample sizes (28, 41 and 25 patients, respectively) and both the control and the Achilles tendinopathy groups were well matched in terms of their baseline characteristics. If proved to be correct, then the absolute presence of neovascularisation could be used as a cardinal diagnostic sign. Unfortunately many other studies have found this not to be the case. Leung et al found no neovascularisation in 14 out of 30 cases (40), Peers et al showed no neovascularisation in 22 out of 25 (43), de Vos et al in 40 out of 67 (30) and Richards et al in 45 out of 55 (44). One possible explanation for these discrepancies might relate to the ultrasound imaging. Ultrasound has a higher positive predictive value of 0.65 and a lower negative predictive value of 0.68 (38), which means there is probable chance of acquiring false- positive and false-negative results. In addition, the studies in question used different ultrasound equipment (models and transducers) and frequency settings as shown in Table 1. Such differences might have contributed to reduced image resolution, thus affecting the results. Despite differing results from de Vos et al (30), Peers et al (43) and Richards et al (44), their methodologies was sound and even

www.sportEX.net

TABLE 1: ULTRASOUND SCANNER MODELS AND TRANSDUCERS USED IN THE NEOVASCULARISATION STUDIES Study

Model Alfredson et al (29) Öhberg et al (28)

Öhberg and Öhberg (42)

de Vos et al (30) Peers et al (43) Leung et al (40) Richards et al (44) Sengkerij et al (45) Siemens Sequoia 512 Siemens Sequoia 512 Siemens Sequoia 512

Unspecified Siemens model

HDI 5000 Philips HDI 5000 Philips HDI 3000 Philips

Unspecified Siemens model

involved extra precautions to minimise obscuring of smaller blood vessels. De Vos et al (30), for example, used minimal probe pressure, and Richards et al (44) and Peers et al (43) set the power Doppler to a pulse repetition frequency of 800–1000 Hz and adjusted the colour gain to exclude signals from normal vessels. From Table 1 it is clear that the studies using colour Doppler reported 100% neovascularisation, which implies that this modality might be more sensitive than power Doppler. Richards et al (44) compared power Doppler to colour Doppler for detecting neovascularisation in Achilles tendinopathy patients. Their results conclusively showed that power

Type of probe (frequency)

Colour Doppler (8–13 MHz)

Colour Doppler (8–13 MHz)

Colour Doppler (8–13 MHz)

Power Doppler (8–13 MHz)

Power Doppler (5–12 MHz)

Power Doppler (12–5 MHz)

Power Doppler (5–12 Mhz)

Power Doppler (8–13 MHz)

Doppler was more accurate, detecting 191 vessels compared to 44 vessels with colour Doppler, leading them to claim that power Doppler was more accurate. Since the publication of their paper in 2005 two of the four of the chosen studies used power Doppler in preference. No other studies have challenged this finding. Leung et al (40), de Vos et al (30), Peers et al (43) and Richards et al (44) have shown with power Doppler that all normal tendons have no neovascularisation, but not all symptomatic Achilles tendinopathy tendons have neovascularisation. These results are comparable across studies because the subjects were similar in age (mid-40s) and the same transducer was used (5–12 MHz) except by de Vos et al.

Symptomatic patients with

neovascularisation 100%

100% 100% 63% 88% 47% 82% 70%

ASSOCIATED NERVES SHOW NEUROGENIC INFLAMMATION. LOCAL ANAESTHETIC IN THE VESSELS REMOVES PAIN AND THEREFORE REMOVING VESSELS MIGHT IMPROVE SYMPTOMS

27

NEOVASCULARISATION IN ACHILLES TENDINOPATHY IS ABNORMAL AND