Cerebral palsy · Review · 2025
Can shockwave help children with cerebral palsy?
This review gathered 12 studies of shockwave in children with cerebral palsy, often alongside physiotherapy or exercise. The studies reported less muscle stiffness after treatment, and some also reported better range of motion, walking, balance and hand use.
Stiffness and movement improved over time- studies in children with cerebral palsy
- 12
- shocks, the most common dose
- 1,500
- months: the longest treatment courses
- 3
Good to know: Studies used different doses and plans, so results were described one by one, not combined.
de Roo EG, Koopman SB, Janssen TW, et al. International Journal of Surgery (London, England), 2025 · Read the paper ↗
How this study worked
What the trials used
- ESWT dose range
- The review reports 500–2,000 shocks (most often 1,500), 4–15 Hz, 0.03–0.32 mJ/mm² or 1.2–4 bar as reported. Do not convert pressure units.
- Targets and co-interventions
- Targets included calf, hamstrings, paravertebral and wrist/elbow flexors. Protocols ranged from one treatment, or three sessions within one week (Gawad), to weekly or twice-weekly courses over weeks or months (for Farhan the text gives one session per month over two months, Table 1 once a week for 2 months); several arms combined ESWT with exercises, conservative care, NDT or orthotic management.
Who took part
Children with cerebral palsy of different ages and types, treated at the calf, back of the thigh, back or arm muscles.
Side effects
Most studies did not report on side effects. Two described the treatment as painless, and one reported no complications.
All the numbers and the full appraisal
What the full paper reports.
Twelve clinical studies of children with cerebral palsy, with varied age, unilateral/bilateral involvement and functional presentation. Included designs and comparator structures differ; outcomes were not pooled.
Resistance to passive movement / muscle tone1 reported outcome
Resistance to passive movement / muscle tone
Study-level scales and measurements as reported
The review reports improvement across studies using MAS and related measures but does not pool results. Gonkova Table 2 gives the placebo sequence separately from the ESWT sequence: MAS 2.70 (SE 0.09) before placebo and 2.63 (0.11) after placebo; before ESWT 2.77 (0.09), after ESWT 2.00 (0.08), at 2 weeks 2.05 (0.07), and at 4 weeks 2.15 (0.07). The post-ESWT values carry superscript a, which the Table 2 footnote defines as P < 0.05 without naming the comparison; Gonkova has no c (between-group difference) marker. Table 2 gives no N for this row; Table 1 lists 25 children, with one ESWT session and one placebo session 4 weeks apart. Amelio MAS was 3.3±0.49 before ESWT and 1.8±0.38 after, with 2.8±0.57 at 12 weeks. These within-study trajectories are not pooled comparative effects.
Analysis details and source
No pooled synthesis was performed because protocols and outcome tools differed. Distinguish within-group change from between-group comparison.
Passive range of motion and joint mobility1 reported outcome
Passive range of motion and joint mobility
Study-level scales and measurements as reported
In Gonkova, pROM was 33.25 (SE 2.20) before ESWT, 47.00 (2.29) after, 46.87 (2.08) at 2 weeks and 44.12 (1.93) at 4 weeks. Wang reported left pROM 18.0±11.6 at baseline, 30.7±9.8 at 1 month and 33.6±11.1 at 3 months with ESWT plus conservative care; comparator values were 18.3±8.6, 20.5±7.6 and 23.4±6.8. Studies and units vary; no pooled estimate. Gonkova pROM means (SE) were 33.13 (2.14) before placebo and 33.05 (2.10) after placebo; before ESWT 33.25 (2.20), after ESWT 47.00 (2.29), at 2 weeks 46.87 (2.08), and at 4 weeks 44.12 (1.93). Table 2 says the review could not calculate effect sizes for Gonkova because raw data were lacking; superscript a (P < 0.05) marks post-ESWT values and is not reinterpreted as a between-group test.
Analysis details and source
No pooled synthesis was performed because protocols and outcome tools differed. Distinguish within-group change from between-group comparison. Table 2 gives no N for Gonkova (Table 1 lists 25 children), and no effect size because raw data were lacking.
Gait, balance, gross motor and hand function1 reported outcome
Gait, balance, gross motor and hand function
Study-level scales and measurements as reported
The narrative reports improvements in gait, balance, GMFM, posture and hand function across small heterogeneous studies; for GMFM, two studies (Lin, Emara) reported significant pre–post differences, while Wang found no time-by-treatment effect after 3 months. Bilek compared NDT+ESWT with NDT alone over 8 weeks; the review calculated between-group effect sizes for its TCMS/TIS/PBS results, and the TUG estimate was 0.65 (95% CI −0.08 to 1.34), an interval including no difference. Other reports mainly provide within-group pre–post effects.
Analysis details and source
No pooled synthesis was performed because protocols and outcome tools differed. Distinguish within-group change from between-group comparison.
Tolerability and adverse effects1 reported outcome
Tolerability and adverse effects
Study-level scales and measurements as reported
The text after Table 2 says two studies (Wang and Elnagger) reported ESWT as non-painful or pain-free without other discomfort and that the remaining studies did not mention side effects. Table 2’s side-effects column also lists Park as “Not painful”, so three studies have a tolerability entry there; the other nine show N/A, which the footer defines as “not available”. Missing reporting is not evidence that no adverse events occurred.
Analysis details and source
No pooled synthesis was performed because protocols and outcome tools differed. Distinguish within-group change from between-group comparison.
As reported in the source review table11 reported outcomes
ASAS change by shock dose (Wardhani et al.)
As reported in the source review table Study-reported measure; units as stated in source
Table 2 reports four dose groups (500, 1000, 1500 or 2000 shocks): mean pre–post / pre–2-week / pre–4-week ASAS changes were Group 1 −1.633/−1.890/−2.060 (4-week marker a); Group 2 −2.271/−2.232 (a)/−2.060 (a); Group 3 −2.121 (a)/−2.460 (a)/−2.456 (a); Group 4 −1.857/−2.271 (a)/−2.232 (a). Raw data were unavailable for the review to calculate effect sizes; the authors conclude shock count did not seem to matter. Table 2 marks only the reported a values as significant; no dose-comparison effect estimate is supplied.
Analysis details and source
These are study-level source-table values, not pooled treatment effects. N/A or calculated pre–post values are not converted into unreported randomized arm contrasts.
Baropodometry after placebo and ESWT (Gonkova et al.)
As reported in the source review table Study-reported measure; units as stated in source
Table 2 reports pressure means (SE): before placebo 50.70 (6.54), after placebo 50.11 (6.43), before ESWT 50.47 (6.61), after ESWT 75.17 (3.42), marked a. Foot-contact means (SE): 80.64 (6.11), 81.82 (6.17), 81.32 (6.14), and 101.58 (5.41), respectively, with the ESWT value marked a. The source does not state the measurement units in this table row; review-calculated effect sizes were not possible due to unavailable raw data.
Analysis details and source
These are study-level source-table values, not pooled treatment effects. N/A or calculated pre–post values are not converted into unreported randomized arm contrasts.
Posture, balance and mobility changes (Bilek et al.)
As reported in the source review table Study-reported measure; units as stated in source
For NDT+ESWT versus NDT alone, Table 2 reports: TCMS total −1.51 (95% CI −2.25 to −0.69); static sitting −3.33 (−4.30 to −2.19); selective movement −1.20 (−1.92 to −0.42); dynamic reaching −2.48 (−3.33 to −1.51); TIS total −1.54 (−2.28 to −0.72); static sitting −1.24 (−1.96 to −0.45); dynamic sitting −0.92 (−1.63 to −0.17); coordination −2.13 (−3.14 to −1.37); Pediatric Balance Scale −3.09 (−4.02 to −2.00); TUG 0.65 (−0.08 to 1.34), whose interval includes zero. These are the review’s calculated between-group pre–post effect summaries; no additional p values are printed in these rows.
Analysis details and source
These are study-level source-table values, not pooled treatment effects. N/A or calculated pre–post values are not converted into unreported randomized arm contrasts.
Gait analysis outcomes with ESWT plus physiotherapy or physiotherapy alone (El‑Shamy et al.)
As reported in the source review table Study-reported measure; units as stated in source
At month 3, MAS was 1.63±0.23 after ESWT+physiotherapy versus 1.86±0.22 after physiotherapy alone; review-calculated within-arm ES were 1.89 (95% CI 0.98–2.69) and 0.96 (0.18–1.69). Gait means±SD and separate within-arm ES: stride length 0.74±0.09 vs 0.50±0.07 (−3.99 [−5.10 to −2.67] vs −1.60 [−2.37 to −0.74]); speed 0.75±0.08 vs 0.60±0.07 (−3.73 [−4.79 to −2.46] vs −1.73 [−2.52 to −0.85]); rhythm 119.93±5.11 vs 125±4.23 (2.30 [1.33 to 3.16] vs 1.22 [0.41 to 1.96]); cycle time 0.65±0.05 vs 0.48±0.07 (−3.75 [−4.81 to −2.48] vs −1.57 [−2.34 to −0.72]); stance 55.93±1.67 vs 50.47±1.85 (−5.61 [−7.02 to −3.90] vs −2.82 [−3.74 to −1.74]); swing 44.07±1.67 vs 49.53±1.85 (5.61 [3.90 to 7.02] vs 2.82 [1.74 to 3.74]). The review's ES values are separate within-arm calculated summaries, not treatment contrasts; units are not consistently stated.
Analysis details and source
These are study-level source-table values, not pooled treatment effects. N/A or calculated pre–post values are not converted into unreported randomized arm contrasts.
Muscle tone, passive ROM and GMFM by side (Wang et al.)
As reported in the source review table Study-reported measure; units as stated in source
For ESWT+conservative care vs conservative care, left MAS: baseline 2.6±1.0 vs 2.5±0.8; 1 month 1.9±0.8 vs 2.4±0.7; 3 months 1.5±1.0 vs 2.1±0.5. Right MAS: 1.9±0.6 vs 1.8±0.7 at baseline; 1.7±0.7 vs 1.8±0.7 at 1 month; 1.2±0.7 vs 1.5±0.7 at 3 months. Left pROM: 18.0±11.6 vs 18.3±8.6 baseline; 30.7±9.8 vs 20.5±7.6 at 1 month; 33.6±11.1 vs 23.4±6.8 at 3 months. Right pROM: 21.9±12.6 vs 17.8±9.9; 32.1±9.4 vs 20.3±9.2; 34.4±10.0 vs 23.6±8.2 at the same timepoints. The review-calculated within-arm ES for the ESWT group were left MAS 0.77 (95% CI 0.27–1.26) at 1 month and 1.10 (0.58–1.60) at 3 months; right MAS 0.31 (−0.17–0.78) and 1.07 (0.55–1.57); left pROM −1.18 (−1.68–−0.65) and −1.37 (−1.89–−0.83); right pROM −0.92 (−1.68–−0.65) and −1.10 (−1.60–−0.58). GMFM was 36.6±28.3 to 52.9±24.4 with ESWT and 40.2±30.3 to 53.1±27.7 in controls at baseline/3 months; the review reports no time-by-treatment effect on GMFM in this study after 3 months.
Analysis details and source
Study-level values and review-calculated within-arm pre–post effects are reproduced as printed; they are not pooled treatment contrasts. Table 2 marks (a, P < 0.05; c, significant difference between groups): left MAS 1.9 and 1.5, left pROM 30.7 and 33.6 and right pROM 32.1 and 34.4 carry a and c; right MAS 1.7 and 1.2 carry a. No missing between-arm estimates are inferred.
Passive ROM and pedobarometry (Amelio et al.)
As reported in the source review table Study-reported measure; units as stated in source
Table 2 means±SD: pROM 20±7 before treatment, 20±6 after placebo, 50±6 after ESWT, 50±7 at 1 week, 40±6 at 4 weeks and 30±8 at 12 weeks. Review-calculated ES values were 4.60 (95% CI 2.95–5.93) after treatment, 4.29 (2.72–5.56) at 1 week and 3.07 (1.80–4.12) at 4 weeks. Peak pressure (kPa) was 20.5±4 before treatment, 22.1±6 after placebo, 99.57±9 after ESWT, 95.06±6 at 1 week, 70.11±11 at 4 weeks and 30.4±9 at 12 weeks; corresponding calculated ES values −11.35 (−14.16–−7.76), −14.62 (−18.19–−10.04) and −5.99 (−6.61–−3.97). Foot contact (cm²) was 40.31±5, 43.15±6, 80.23±7, 70.12±5, 68.09±8 and 48.23±6 at these same points; calculated ES −6.56 (−8.30–−4.37), −5.96 (−7.57–−3.94) and −4.16 (−5.41–−2.63). Table 2 gives no 12-week ES.
Analysis details and source
Study-level values and review-calculated within-arm pre–post effects are reproduced as printed; they are not pooled treatment contrasts. In Table 2, pROM, peak pressure and foot contact after ESWT, at 1 week and at 4 weeks carry b (P < 0.01); the 12-week values carry no mark. No missing between-arm estimates are inferred.
Upper-limb tone and hand function (Farhan et al.)
As reported in the source review table Study-reported measure; units as stated in source
The source review reports pre–post differences (mean ± a dispersion value the table does not label as SD or SE) for ESWT plus exercise versus exercise alone: elbow MAS 1.28±0.16 vs 0.41±0.07; wrist MAS 0.94±0.12 vs 0.38±0.07; and Modified House Functional Classification (MHC) 1.50±0.16 vs 0.56±0.13. Values were marked b for the ESWT group. Review authors state effect sizes could not be calculated because data were insufficient. No between-group p value or comparative estimate is supplied in this row.
Analysis details and source
Study-level values and review-calculated within-arm pre–post effects are reproduced as printed; they are not pooled treatment contrasts. Source markers a/b/c are preserved when stated, and no missing between-arm estimates are inferred.
Electrophysiology and gait outcomes (Gawad et al.)
As reported in the source review table Study-reported measure; units as stated in source
Table 2 baseline→1-week means±SD, ESWT+exercise vs exercise: H/M ratio 3.75±1.08→1.95±0.60 vs 4.00±1.20→3.99±1.20 (review ES for the ESWT group 2.06, 95% CI 1.13–2.88); gait speed 0.67±0.04→0.90±0.06 vs 0.68±0.02→0.68±0.02 (ES −4.51, −5.71 to −3.07); cadence 118.20±2.33→117±2.20 vs 119.60±2.75→119.40±2.77 (ES 0.53, −0.21 to 1.24); stride length 0.70±0.02→0.96±0.05 vs 0.72±0.45→0.73±0.44 (ES −6.83, −8.47 to −4.82); single-limb support 0.36±0.12→0.40±0.20 vs 0.36±0.13→0.36±0.14 (ES −0.24, −0.95 to 0.48); double-limb support 0.26±0.04→0.21±0.04 vs 0.27±0.04→0.27±0.04 (ES 1.25, 0.44 to 2.00). Dorsiflexion: initial contact −3.87±1.85→7.20±1.87 vs −3.53±1.51→−3.33±1.72 (ES −5.95, −7.42 to −4.16); midstance 5.13±2.20→12.13±2.42 vs 5.53±1.81→5.67±2.02 (ES −3.03, −3.98 to −1.91); stance phase 9.53±1.81→16.20±1.82 vs 10.80±2.08→11.00±2.00 (ES −3.03, −3.98 to −1.91); mid-swing −3.93±3.10→7.00±1.36 vs −2.47±3.11→−2.27±3.24 (ES −4.57, −5.77 to −3.11). The table gives within-arm calculated ES for the ESWT group only, not treatment contrasts.
Analysis details and source
These are reported study-level endpoint values and the review’s within-arm calculated ES, not a pooled treatment contrast. Interpret source signs and units as printed.
Strength, gross motor function and balance (Emara et al.)
As reported in the source review table Study-reported measure; units as stated in source
Table 2 reports baseline→post means±SD for ESWT+exercise vs exercise. Biodex peak eccentric torque 60°: 6.63±1.14→5.03±1.52 vs 7.53±1.87→6.18±1.35; review ES 1.19 (95% CI 0.44–1.89) vs 0.83 (0.11–1.51). At 180°: 5.98±1.07→3.29±1.35 vs 5.41±1.09→5.17±0.92; ES 2.21 (1.31–3.00) vs 0.24 (−0.44–0.91). Threshold 60°: 12.98±3.33→17.18±2.18 vs 11.88±2.83→14.79±1.79; ES −1.49 (−2.21 to −0.70) vs −1.23 (−1.93 to −0.47). Threshold 180°: 8.60±2.56→14.87±3.57 vs 8.80±2.37→11.35±2.33; ES −2.02 (−2.79 to −1.15) vs −1.09 (−1.78 to −0.34). GMFM-D: 29.55±1.58→37.23±1.67 vs 28.86±3.59→34.96±1.83; reported ES −4.72 (95% CI −5.90 to 3.33 as printed; the upper bound is inconsistent with the estimate, and the interval is not charted) vs −2.14 (−2.93 to −1.25). GMFM-E: 42.96±2.29→58.56±3.58 vs 42.86±3.95→50.78±4.19; ES −5.19 (−6.45 to −3.69) vs −1.95 (−2.71 to −1.09). Single-leg standing: 2.46±0.37→6.54±0.49 vs 2.17±0.52→5.17±0.88; ES −9.40 (−11.46 to −6.89) vs −4.15 (−5.23 to −2.88). MAS was reported as medians: study group 1+ (1–1+) to 1 (1–1+), control 1+ (1–1+) to 1 (1–1+); source says ES could not be calculated. SMC medians were 0.6250 (0–2) to 1.7394 (0–2) and 0.4943 (0–2) to 1.0573 (0–2), respectively; ES not calculable.
Analysis details and source
These are reported study-level endpoint values and the review’s within-arm calculated ES, not a pooled treatment contrast. Interpret source signs and units as printed.
MAS, pROM and GMFM results (Park and Lin et al.)
As reported in the source review table Study-reported measure; units as stated in source
Park Table 2: study MAS 2.3±0.4 at baseline, 1.2±0.4 after first session, 1.0±0.4 after third, 1.1±0.5 at 4 weeks; review ES 2.75 (95% CI 1.02–4.06), 3.25 (1.35–4.65), and 2.65 (−0.95–3.94). Control MAS 2.5±0.7, then 1.5±0.5 after first ESWT, 2.0±0.7 after third placebo, 2.3±0.8 at 4 weeks; ES 1.64 (0.23–2.80), 0.71 (−0.50–1.82), 0.27 (−0.89–1.38). Study pROM 6.6±2.5→13.6±2.5 after first, 22.5±2.7 after third, 19.1±3.7 at 4 weeks; ES −2.80 (−4.12–−1.05), −6.11 (−8.16–−3.12), −3.96 (−5.50–−1.80). Control 5.8±2.0→13.3±3.7 after first ESWT, 7.9±1.3 after third placebo, 7.3±1.8 at 4 weeks; ES −2.52 (−3.79–−0.86), −1.25 (−2.37–0.07), −0.79 (−1.90–0.44). Lin Table 2 (review ES given for the ESWT group only): MAS hamstrings 4.56±0.77→2.73±0.57 at 2 weeks and 2.26±0.71 at 1 month, versus control 4.72±0.59→3.39±0.51 and 2.96±0.85; review ES 2.70 (2.09–3.26) and 3.11 (2.45–3.70). Triceps surae MAS 4.49±0.39→2.66±0.34 and 2.20±0.33 vs 4.68±0.53→3.41±0.54 and 3.01±0.61; ES 5.00 (4.10–5.81) and 6.34 (5.25–7.31). GMFM 54.72±10.33→73.6±12.1 and 75.9±6.6 vs 55.67±9.36→62.7±9.8 and 66.2±10.3; ES −1.68 (−2.15–−1.17), −2.44 (−2.98–−1.86). Foot area 50.43±6.82→60.98±5.96 and 66.27±5.71 vs 50.88±6.73→54.34±6.01 and 56.77±5.85; ES −1.65 (−2.12–−1.14), −2.52 (−3.06–−1.93).
Analysis details and source
These are reported study-level endpoint values and the review’s within-arm calculated ES, not a pooled treatment contrast. Interpret source signs and units as printed.
Balance, gait and EMG across three groups (Elnagger et al.)
As reported in the source review table Study-reported measure; units as stated in source
Table 2 compares Group1 ESWT+training, Group2 SOM+training, and Group3 ESWT+SOM+training. H/M ratio most affected side: baseline/post 0.36±0.06/0.32±0.05, 0.33±0.04/0.31±0.05, 0.34±0.05/0.28±0.03; within-arm ES 0.72 (95% CI 0.03–1.38), 0.44 (−0.27–1.13), 1.46 (0.71–2.14); between-group p=.005. Less affected side: 0.34±0.05/0.30±0.04, 0.31±0.04/0.28±0.03, 0.34±0.05/0.28±0.03; ES 0.88 (0.18–1.55), 0.85 (0.11–1.55), 1.21 (0.50–1.88); p=.029. PBS: 35.31±3.42/37.78±4.11, 37.28±4.61/39.24±3.79, 36.58±3.51/41.32±4.14; ES −0.65 (−1.31–0.03), −0.46 (−1.15–0.25), −1.29 (−1.96–−0.57); p=.025. Gait hip angle: 36.83±2.48/35.00±2.57, 35.18±3.40/33.25±2.77, 34.95±3.27/32.58±2.71; ES 0.72 (0.04–1.38), 0.62 (−0.10–1.32), 0.79 (0.11–1.43); p=.025. Knee angle: 35.22±3.39/33.17±2.92, 34.00±2.16/31.69±3.05, 33.37±3.17/30.79±2.42; ES 0.65 (−0.04–1.30), 0.87 (−0.13–1.58), 0.91 (0.23–1.56); p=.041. Ankle plantar-flexion angle: 33.44±2.15/31.89±2.65, 32.56±3.03/30.65±2.31, 31.95±2.50/29.74±2.23; ES 0.64 (−0.04–1.30), 0.71 (−0.02–1.40), 0.93 (0.25–1.58); p=.031. Velocity: 0.89±0.10/0.94±0.11, 0.92±0.13/0.95±0.14, 0.93±0.14/1.05±0.17; ES −0.48 (−1.13–0.20), −0.22 (−0.91–0.48), −0.77 (−1.41–−0.10); p=.032. Cadence: 109.22±9.74/104.23±7.44, 102.87±12.66/98.69±10.44, 105.37±11.57/97.05±6.41; ES 0.58 (−0.10–1.23), 0.36 (−0.35–1.05), 0.89 (0.21–1.54); p=.027. The table’s marked ES/test semantics are retained; these within-arm effects are not pairwise treatment effects.
Analysis details and source
These are reported study-level endpoint values and the review’s within-arm calculated ES, not a pooled treatment contrast. Interpret source signs and units as printed.
Modified Ashworth Scale after ESWT plus physiotherapy or physiotherapy alone (El-Shamy et al.)1 reported outcome
Modified Ashworth Scale after ESWT plus physiotherapy or physiotherapy alone (El-Shamy et al.)
Modified Ashworth Scale MAS points
Table 2 reports MAS falling from 2.34±0.48 to 1.63±0.23 with ESWT plus physiotherapy (review-calculated within-arm ES 1.89, 95% CI 0.98–2.69) and from 2.27±0.56 to 1.86±0.22 with physiotherapy alone (within-arm ES 0.96, 95% CI 0.18–1.69). These are before–after changes within each arm; they are not a between-group treatment effect.
Analysis details and source
The standardized effects are values calculated by the review authors from within-arm data; they must not be interpreted as ESWT-versus-control contrasts.
How treatment was delivered
The review reports 500–2,000 shocks (most often 1,500), 4–15 Hz, 0.03–0.32 mJ/mm² or 1.2–4 bar as reported. Do not convert pressure units.
Targets included calf, hamstrings, paravertebral and wrist/elbow flexors. Protocols ranged from one treatment, or three sessions within one week (Gawad), to weekly or twice-weekly courses over weeks or months (for Farhan the text gives one session per month over two months, Table 1 once a week for 2 months); several arms combined ESWT with exercises, conservative care, NDT or orthotic management.
Table 2 uses “N/A” (its footer defines N/A as “not available”) for Wardhani, Gonkova, Bilek, El‑Shamy, Amelio, Farhan, Gawad, Emara and Lin; these entries do not establish that harms were actively assessed or absent. Park explicitly reports “Not painful.” Elnagger explicitly reports “Pain-free, no other discomfort.” Wang separately reports no complications/adverse effects and no complaints directly related to rESWT. These affirmative tolerability statements are distinct from N/A and from studies with no safety details. The review does not provide a complete pooled adverse-event count.
Results were not pooled because outcome measures, ESWT protocols and co-interventions varied.
PEDro ratings ranged from poor to excellent; one study poor, two fair, eight good and one excellent.
The review lacked access to Embase, Web of Science and CINAHL; relevant studies may have been missed.
Tolerability was sparsely reported: Table 2 gives tolerability statements for three studies (Park, Wang, Elnagger), though the text after it names two (Wang, Elnagger); the other nine entries are N/A (not available).
Many reported effect sizes were calculated from pre–post data; they should not be confused with controlled between-group effects.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Results were not pooled because outcome measures, ESWT protocols and co-interventions varied.
- PEDro ratings ranged from poor to excellent; one study poor, two fair, eight good and one excellent.
- The review lacked access to Embase, Web of Science and CINAHL; relevant studies may have been missed.
- Tolerability was sparsely reported: Table 2 gives tolerability statements for three studies (Park, Wang, Elnagger), though the text after it names two (Wang, Elnagger); the other nine entries are N/A (not available).
- Many reported effect sizes were calculated from pre–post data; they should not be confused with controlled between-group effects.
Source-linked critical reading
Systematic clinical review
Technical source record
Complete full-text XML, both raw tables and the figure caption read; primary studies were not independently acquired.
Small studies suggest ESWT may help muscle stiffness and some movement outcomes in children with CP, often alongside rehabilitation. The evidence is varied and does not yet identify the best protocol or prove lasting functional benefit.
What effects and tolerability of ESWT have been reported in children with cerebral palsy across body function, activity and participation?
Systematic review of 12 heterogeneous clinical studies in children with CP. The review did not pool study results because interventions and outcomes differed. PEDro quality scores were reported for included studies.
Children with varied CP distributions/severity, ages and mobility. Table 1 includes randomized trials, case-control/pilot and placebo-controlled studies; sample units and baseline characteristics differ substantially.
Most protocols used 1,500 shocks; reported range 500–2,000, 4–15 Hz, 0.03–0.32 mJ/mm² or 1.2–4 bar (units as reported). Targets included calf, hamstrings, paraspinal and wrist/elbow flexor muscles. Sessions ranged from a single exposure, or three sessions in one week, to weekly or twice-weekly courses lasting up to 3 months; co-care included exercise, conservative physiotherapy, NDT or orthoses.
Comparisons varied: ESWT alone versus placebo, ESWT added to exercise/conservative care, or active rehabilitation. Wardhani compared four shock doses without a group that did not receive ESWT, and many reported effect sizes are within-arm pre–post measures, limiting causal interpretation.
Across 12 heterogeneous study reports, Table 2 gives endpoint trajectories, pre–post changes and review-calculated within-arm ES; there is no pooled estimate. Source-linked records preserve Wardhani dose/timepoint ASAS; Gonkova placebo/ESWT tone, ROM and baropodometry; Bilek posture/balance contrasts; and gait, GMFM, upper-limb, balance and EMG measures from the remaining studies. Many reported ES are within-arm, not treatment contrasts. Gonkova’s placebo period is kept distinct: MAS 2.70 (SE 0.09) before and 2.63 (0.11) after placebo; pROM 33.13 (SE 2.14) before and 33.05 (2.10) after placebo. Table 2 gives no effect size for Gonkova because raw data were lacking.
Bilek reports source-calculated between-group pre–post summaries for TCMS/TIS/PBS, while TUG 0.65 (95% CI −0.08 to 1.34) includes zero. Elnagger reports three-group tests for H/M ratio, PBS, gait angles/velocity/cadence. Other numerical ES are often within-arm and cannot establish ESWT superiority. The review did not identify an optimal dose. For Gonkova, placebo-period and ESWT-period values are listed as separate sequences; Table 1 lists 25 children, and the review gives no effect size because raw data were lacking.
Table 2 lists N/A for nine reports, explicitly “Not painful” for Park and “Pain-free, no other discomfort” for Elnagger. Wang reports no complications/adverse effects or rESWT-related complaints. These statements are not interchangeable; incomplete N/A reporting is not evidence of zero events.
No pooled analysis due to clinical/outcome heterogeneity; small studies and varied protocols; study quality ranged poor to excellent by PEDro (one poor, two fair, eight good, one excellent). The review also reports restricted database access and Level B evidence across included studies.
ESWT may be considered an adjunct under study for selected children with CP, but the evidence does not establish a standard dose, durable participation gains, or reduced need for surgery. Outcome scales and co-interventions differ.
Conduct adequately powered pediatric trials with matched rehabilitation, prespecified patient-important outcomes, clear CP phenotype and functional level, standardized ESWT dose/target, systematic adverse-event monitoring and follow-up beyond treatment.
Read the original article.
The effects of extracorporeal shock wave therapy in children with cerebral palsy: a systematic review.
de Roo EG, Koopman SB, Janssen TW, Aertssen WFM. · International journal of surgery (London, England) · 2025
Open the publication record or available full-text files.
Checked against the full paper · Clinical review: Dr. Marcus Yu Bin Pai, · How we review →