Summary of findings for kinematic studies (n=13)
| Study | Findings | Narrative |
| An et al 201529 | No differences in the ASI (%) of PF, mean loading rate and peak loading rate between the healthy rowers and rowers with a history of LBP (*no mean±SD data provided, p=0.448–0.722, Hedges’ g=0.162–0.310). Subgroup analysis found no differences between healthy scullers versus scullers with a history of LBP or healthy sweep rowers versus sweep rowers with a history of LBP (p=0.194–0.855, Hedges’ g=0.203–0.518). | Asymmetry not significant link to LBP. |
| Buckeridge et al 201213 | As work intensity increased, hip and knee ROM ↓ across all groups (*no mean±SD data provided, p<0.01). Knee ROM demonstrated no differences across all groups (p=0.21) however, elite rowers appeared to exhibit greater knee extension at the finish position than novice rowers (*no mean±SD data provided). There was a trend for hip ROM to ↓ from elite to novice rowers, thus hip ROM may have the potential to distinguish between groups (elite rowers=right hip ROM 97.2°±10.6°, left hip ROM 92.7±9.0°; club rowers=right hip ROM 94.9°±10.7°, left hip ROM 93.2°±8.2°; novice rowers=right hip ROM 90.6°±8.4°, left hip ROM 87.2°±8.1°; p=0.08). As work intensity increased, LP flexion at the catch ↑ in all groups (*no mean±SD data provided, p<0.01). Regression analysis indicated that both the hip and knee ROM ASI (%) predicted LP flexion (p<0.05) however, hip ROM ASI showed a better predictive relationship with LP flexion (r2=0.35) compared with knee ROM asymmetry (r2=0.08). | As work intensity increased, hip and knee ROM ↓ and all groups move more through the pelvis, however elite rowers control movement better. Leg asymmetry does not influence spine and pelvic movement. |
| Fletcher et al 201538 | Significant difference between Pose and Rosenberg rowing techniques in SR (Pose=37±1 strokes/min; Rosenberg=32±1 strokes/min), SL (Pose=1.1±0.1 m; Rosenberg=1.3±0.1 m), oar handle impulse (Pose=233±21 N.s; Rosenberg=266±24 N.s) and trunk extension (Pose=−14°±0.5°; Rosenberg=−29°±1.0°). | Change in kinematics with Pose rowing technique ↑ SR, ↓ SL, ↓ oar handle impulse and ↓ trunk extension. |
| Holt et al 200316 | As work duration increased, maximum LS flexion ↑ (−0.7°±1.4° to −2.3°±2.3°, p=0.001) and extension ↑ (36.3°±4.3° to 38.0°±8.5°, p=0.04). As work duration increased, gradient of force production for drive phase 1 ↓ (96.9°±29.2° to 86.9°±24.1°, p=0.01) and drive phase II ↓ (66.1°±19.8° to 59.9°±22.3°, p=0.01). | As work duration increased, there was a change in LP kinematics ↑ lumbar movement and ↓ hip flexion. |
| MacManus et al 201328 | No difference in LP kinematics during an incremental test for the healthy group and history of LBP group (*no mean±SD data provided, p=0.541), although there was a trend for flexion to ↑ at the last stage of the test (*no mean±SD data provided). | No change in LP kinematics with fatigue in the healthy group and history of LBP group. |
| McGregor et al 2002a35 | At the catch, the current LBP group had a loss of angulation at L5/S1 (into posterior pelvic rotation) (2.8°±5.5° in the current LBP group vs 4.8°±1.2° in the history of LBP group vs 7.5°±1.3° in the healthy group, p<0.05) and L1/L2 (*no mean±SD data provided, p<0.01). At the finish, the current LBP group rotate sacra and pelvis posteriorly compared with the healthy rowers who maintained their lumbar spines in a neutral position (*no mean±SD data provided, p<0.05). | Rowers with current LBP or a history of LBP move more through their LS and less through hips/pelvis compared with healthy rowers. |
| McGregor et al 2004b27 | As work intensity increased, anterior pelvic rotation ↓ at the catch (0.6°±4.3° to −6.3°±6.6°, p<0.001) and there was a non-significant trend for posterior pelvic rotation to ↑ (−44.5°±13.3° to −47.8°±11.7°). | As work intensity increased, anterior pelvic rotation ↓ at the catch. |
| McGregor et al 200526 | As work intensity increased, there was a non-significant trend for SL to ↓ (136.5 cm±6.4 cm to 130.6 cm±8.1 cm), anterior pelvic rotation to ↓ at the catch (13.3°±6.1° to 8.5°±4.9°) and for posterior pelvic rotation to ↑ at the finish (−13.5°±8.8° to −19.7°±8.4°). Maximum posterior pelvic rotation occurred later in the stroke as work intensity increased (33.3%±2.8% during step 1 to 40.2%±2.7% during step 6, p<0.0001). There was a non-significant trend for lumbar rotation to ↓ (26.0°±5.0° to 22.1°±7.4°) and trunk extension to ↑ at the finish (−13.8°±10.5° to −22.4°±7.1°) as work intensity increased. | As work intensity increased, there was a trend for SL to ↓, anterior pelvic rotation to ↓ at the catch and for posterior pelvic rotation to ↑ at the finish. This may be a reflection of poor trunk control and muscle endurance. |
| McGregor et al 200725 | Over a 2-year period, PF ↑ (746±77 n to 792±67 n, p<0.05), SL ↑ (141±0.7 cm to 156±0.5 cm, p<0.05), anterior pelvic rotation ↑ at the catch (12.5°±5.8° to 17.4°±3.5°, p<0.001), thigh extension ↑ at the finish (2.2°±2.2° to 5.7°±5.6°, p<0.05) and there was a non-significant trend for rowers to achieve anterior pelvic rotation earlier in the stroke (88.6%±8.2% to 83.9±10.6%, p=0.09) during the final stage of an incremental test. | Over a 2-year training period, trunk endurance improved LP ratio allowing equal amounts of lumbar and pelvic rotation. Training is important to develop good LP patterning. |
| Steer et al 200622 | Significant difference in thigh rotation at the finish on the WaterRower ergometer compared with Concept 2 ergometer at a SR of 18–20 (−21.9°±4.2° vs −13.7°±4.6°, p=0.001) and SR of 28–30 (−20.6°±5.4° vs −15.7°±4.2°, p=0.01). Significant difference in pelvic rotation on the WaterRower ergometer compared with Concept 2 ergometer at a SR of 18–20 (*no mean±SD data provided, p=0.03) but not at a SR of 28–30 (*no mean±SD data provided, p=0.32). The mean angle of the pelvis at the catch was 2.4° on the Concept 2 ergometer and 5.4° on the WaterRower. | ↓ anterior pelvic rotation at the catch on the WaterRower compared with Concept 2 and the reduced thigh rotation at the finish suggests the rowers were not fully straightening their legs on the WaterRower. |
| Strahan et al 201121 | Sweep rowing had greater lateral bend throughout the stroke (*no mean±SD data provided, p<0.05), which was predominately due to movement of the upper lumbar and lower thoracic regions. Furthermore, sweep rowing displayed a greater magnitude of axial rotation at the catch (created at the pelvis) (*no mean±SD data provided, p<0.05). Both sweep rowing and scull rowing showed values close to end range flexion for the lower lumbar spine at the catch and early drive phases. No difference was evident in lateral bend or axial rotation values for the lower lumbar region (*no mean±SD data provided, p>0.05). | Sweep rowing and sculling differs primarily in axial bending, mostly at the thoracic spine. |
| Wilson et al 201219 | Frontal plane LS angular displacement was present during ergometer rowing (4.6°–8.7° at L3) and ↑ from the first to the last stage of an incremental test (mean increase=4.18°±1.94°, 95% CI=2.9° to 5.3°, p<0.001). Blood lactate (as a fatigue measure) also ↑ during the test (from 1.2±0.2 to 9.0±2.1 mmol/L) but the ↑ in SR (from 18 to 30 strokes/min) was a better predictor of the ↑ in angular displacement (p<0.0001, r2=0.335). | Movement of the LS in the frontal plane exists during ergometer rowing and ↑ with fatigue and SR. This may be of importance to the development of LBP. |
| Wilson et al 201315 | Maximum LS flexion ↑ during the incremental test and was significantly greater on the ergometer (4.4° c±0.9° c hange) compared with the boat (1.3° c±1.1° c hange) (3.1° difference, 95% CI=0.3° to 5.9°, p=0.035). Compared with the voluntary ROM, there is an ↑ of 11.3% (ergometer) and 4.1% (boat). | Sagittal LS motion ↑ during an incremental test to greater than full standing flexion. Greater ↑ on an ergometer than in a boat. |
↓, decrease (s); ↑, increase (s); ASI, asymmetry index; LBP, low back pain; LP, lumbo-pelvic; LS, lumbar spine; PF, peak force; ROM, range of motion; SL, stroke length; SR, stroke rate.