Table 5 Results table

Bolton et al. 2018 UK [3]

USS features of: Greater trochanteric bursitis, Gluteal tendinopathy, Greater trochanteric cortical irregularity, Gluteus Medius bursitis

Nil found

Presence of Greater Trochanteric Cortical Irregularity (GTCI) on USS was, in unadjusted models, associated with a successful outcome in pain with activity at 6 weeks (OR 3.09 [95%CI, 1.06–9.01]; P = 0.04). When the models were adjusted for age, BMI, gender, and evidence of OA the findings were no longer significant (OR 1.21 [95%CI, 0.39–3.73]; P = 0.74)

GMB was found also shown to be associated with successful outcomes at 6 weeks at rest, in unadjusted models, (OR 4.49 [95%CI, 1.10–18.36]; P = 0.03) and with activity (OR 4.94 [95%CI, 1.07–22.79]; P = 0.03). Results lost significance when adjusted with (OR 6.10 [95%CI, 0.98–37.91]; P = 0.05) and (OR 6.10 [95%CI, 0.98–37.91]; P = 0.05) with pain at rest and activity respectively

No association between GTCI and GMB findings on USS and successful response to injection with reduction in pain score at rest or with activity at 6 months and 12 months in both adjusted and unadjusted models (P < 0.06)

No association between GTB or Gluteal tendinopathy and successful response to injection at rest or with activity at 6 weeks, 6 months, and 12 months in both unadjusted and adjusted models (P < 0.10)

Cohen et al. 2009 USA [13]

Blind Vs guided injections, Intra-bursal Vs extra bursal injection location, Patient characteristics such as: age, sex, obesity, duration of pain, opioid use, baseline pain at rest/activity and Oswestry score

Nil found

Patient characteristics

No differences in mean age (years) between responders 54.6 ± 14.8 and non-responders 55.9 ± 14.4 (P = 0.74)

No difference between patient sex with female % at 83% in non-responders and 89% in responders (P = 0.72)

No association between obesity and responder status 11/36 (31%) of patients were obese in the non-responder group compared to 6/28 (21%) in the responder group (P = 0.57)

No association between baseline pain intensity at rest (mean NRS) in responders 4.7 ± 2.6 compared to non-responders 4.8 ± 2.6 (P = 0.26)

No differences in pre-procedure Oswestry disability index scores between responders 42.8 ± 12.6 and non-responders 36.6 ± 14.3 (P = 0.46)

No difference between responders and non-responders in pain duration with 5.4 ± 4.5 years and 2.7 ± 2.4 years respectively (P = 0.40)

Injection guidance

No differences in mean pain at one month with rest, fluoroscopically guided 2.7 ± 2.4 and landmark guided 2.2 ± 2.4 (P = 0.41), and with activity 5.0 ± 2.9 compared to 4.0 ± 2.6 (P = 0.16)

No differences at three months with fluoroscopically guided mean pain at rest 1.9 ± 1.7 and landmark guided 2.6 ± 2.5 (P = 0.34) and with activity 4.7 ± 2.8 compared to 4.8 ± 2.6 (P = 0.90)

There were also no differences in overall success rating (P = 0.38), Oswestry disability index (P < 0.68), global perceived effect (P = 0.80), and analgesia use (P = 0.60)

Injection location

No differences between injection location between intra-bursal or extra-bursal in overall success (P = 0.72), mean pain intensity at rest and with activity at 1 (P < 0.53) and 3 months (P < 0.31). Also no differences in Oswestry score (P < 0.45), global perceived effect (P = 0.78) and analgesia use (P = 0.14)

Farmer et al. 2010 USA [19]

Age, diagnosis pre surgery, approach of hip, leg length discrepancy, offset and lateralisation of replacement and contralateral THR

Nil found

No differences in responder status by age (P = 0.1), diagnosis pre surgery (P = 0.4), approach of hip surgery (P = 0.4), contralateral total hip replacement (P = 0.7), leg length discrepancy (P = 0.1), offset (P = 0.3) and lateralisation (P = 0.6)

Jarlborg et al. 2021 Switzerland [33]

Positive ‘triple test’ = patients with positive tests of Lequesne, Patricks and 30 s single leg stance

Nil found

No significant differences in outcomes between patients with a triple test + ve or -ve (P = 0.14)

Mao et al. 2020 USA [42]

Location of injection: Non- bursal, Subgluteus Medius bursa, and trochanteric bursa

GTB injections maintained the pain reduction between immediate and 1 week with a non-significant change of 0.55 (P = 1.0). Non-bursal and sub-GMB pain scores increased between immediately post injection and 1 week with 1.50 (P = 0.012) and 0.86 (P = 0.014) respectively

No differences between injection location between sub-GMB (P = 0.46), GTB (P = 1.0) and non-bursal (P = 1.0) immediately post procedure or at 1 week (P = 1.0)

McEvoy et al. 2013 USA [43]

Injection location of GMB Vs GTB

GTB injections resulted in statistically significant reduction in median pain scores of 3.0 (P < 0.01) compared to sub-GMB injections with non-significant reductions of 0 (P = 0.44) and a significant difference between the two injection locations. (P < 0.01)

Nil

Mitchell et al. 2018 USA [46]

Ultrasound guided Vs landmark guided

Significant differences between groups in pain score at 6 months; mean pain scores with USS 3.9 ± 2.0 landmark guided 5.5 ± 2.6 (P = 0.036)

No significant difference in mean pain scores at 2 weeks post CSI: USS 1.3 ± 1.9 and landmark 2.2 ± 2.5 (P = 0.61)

No differences in procedural pain; USS 3.3 ± 2.3 and landmark 4.9 ± 4.1 (P = 0.20)

No differences in time to next intervention (months) USS 8.7 ± 2.9 and landmark 8.3 ± 3.8 (P = 0.75)

No differences in duration of therapeutic effect (months): USS 4.7 ± 1.4 and landmark 4.1 ± 2.9 (P = 0.48)

Park et al. 2016 Rep of Korea [54]

Patient characteristics: Age, sex, pain duration, co-morbidity, BMI, presence of bursitis, tendinosis, enthesitis or partial/full thickness tears on USS

Worse outcomes associated with having a co-existing MSK condition of knee OA (OR 0.329 [95%CI, 0.128–0.848]; P = 0.01) and Sacro-Illiac Joint (SIJ) or lumbar spine pain (OR 0.304 [95%CI, 0.118 – 0.783]; P = 0.01)

No association between pain reduction and age (P = 0.661), gender (P = 0.558), BMI (P = 0.172) and pain duration (P = 0.187)

No association between pain reduction and imaging findings of tendinosis OR 3.17 [95%CI, 0.89–11.33], bursitis OR 4.20 [95%CI, 0.87–20.33], partial tear OR 2.94 [95%CI, 0.76–10.90], full thickness tear OR 3.27 [95%CI, 0.85–12.63] and enthesopathy OR 3.82 [95%CI, 0.81–18.03]