The validity and reliability of a dynamic neuromuscular stabilization-heel sliding test for core stability

Abstract

BACKGROUND:

Core stabilization plays an important role in the regulation of postural stability. To overcome shortcomings associated with pain and severe core instability during conventional core stabilization tests, we recently developed the dynamic neuromuscular stabilization-based heel sliding (DNS-HS) test.

OBJECTIVE:

The purpose of this study was to establish the criterion validity and test-retest reliability of the novel DNS-HS test.

METHOD:

Twenty young adults with core instability completed both the bilateral straight leg lowering test (BSLLT) and DNS-HS test for the criterion validity study and repeated the DNS-HS test for the test-retest reliability study. Criterion validity was determined by comparing hip joint angle data that were obtained from BSLLT and DNS-HS measures. The test-retest reliability was determined by comparing hip joint angle data.

RESULTS:

Criterion validity was (ICC ${}_{2,3}$ ) $=$ 0.700 ( $p<$ 0.05), suggesting a good relationship between the two core stability measures. Test-retest reliability was (ICC ${}_{3,3}$ ) $=$ 0.953 ( $p<$ 0.05), indicating excellent consistency between the repeated DNS-HS measurements.

CONCLUSIONS:

Criterion validity data demonstrated a good relationship between the gold standard BSLLT and DNS-HS core stability measures. Test-retest reliability data suggests that DNS-HS core stability was a reliable test for core stability. Clinically, the DNS-HS test is useful to objectively quantify core instability and allow early detection and evaluation.

Keywords

Core instability criterion validity test-retest reliability dynamic neuromuscular stabilization-heel sliding test rehabilitation

1. Background

Core stabilization plays an important role in the regulation of postural stability, which is required for coordinated dynamic movement. Core stability is achieved by coordinated co-activation of the transverse abdomen (TrA), internal obliques, external obliques, rectus abdomens, pelvic floor muscles, multifidus, and diaphragm. Unconscious or subconscious activation of these muscles generates intra-abdominal pressure or tension in conjunction with breathing, this results in stabilized and upright neutralized spinal posture during dynamic movement [1]. Impaired core stabilization may result in low back pain (LBP) or other musculoskeletal spinal injuries during dynamic lifting movements. Empirical evidence suggests that in the absence or lack of core stabilization, neuromuscular stabilizing forces are unbalanced, resulting in unevenly distributed spinal loads [1]. This results in LBP and spinal alignment deformities (i.e., lordosis, scoliosis, and kyphosis), and may explain why individuals with core instability are prone to spinal pathology. A valid and reliable core stabilization test that accurately and consistently quantifies core instability is necessary.

Core instability is conventionally measured by various methods, including the bilateral straight leg lowering test (BSLLT), formal test [3], and global muscular endurance test (GMET) [4]. BSLLT is widely used to measure abdominal muscle strength. The test involves gradually extending straight legs from 90 ${}^{\circ}$ hip flexion while maintaining a target pressure (40 mmHg $\pm$ 4–10 mmHg) recorded by a pressure biofeedback unit (PBU) in the supine position. The hip joint angle is determined when pressure drops below 10 mmHg (5–7). The formal test, less commonly used, primarily measures lumbar segmental stabilization in the prone position as an individual attempts to maintain the target pressure in PBU after he or she pulls the abdominal muscles into the lumbar spine to selectively activate TrA muscles. GMET is a more advanced core stabilization test for athletes. It is intended to measure overall “trunk” or core endurance and strength performance for the lumbopelvic-hip core muscle chain. GMET encompasses co-activation of the deep and superficial (rectus) abdominal muscles, multifidus, gluteus maximus and medius, iliopsoas, rectus femoris, and other associated pelvic floor core muscles during different positional tests [4]. Surprisingly, only one study demonstrated high inter-tester reliability in GMET measurements ( $r=$ 0.81–0.95) [8]. As with other core assessments, however, no reliability coefficients have been reported for the formal test to date. Furthermore, the majority of individuals with acute or subacute LBP may not be able to perform these challenging core stability tests because of inability to maintain the target pressure and insufficient neuromuscular activation of the diaphragm and deep core stabilizers to counterstabilize excessive anterior shearing force either during BSLLT or GMET [9, 10, 11].

To overcome shortcomings associated with pain and severe core instability during conventional core stabilization tests, we recently developed the dynamic neuromuscular stabilization-based heel sliding (DNS-HS) test. The DNS-HS test is conceptually derived from the DNS technique, which involves subconscious co-activation of the diaphragm, TrA, and multifidus with other superficial core muscles to provide upright stability of the lumbopelvic system in a supine position [1].

While the conventional BSLLT or GMET tests were purported to measure abdominal muscle strength and core endurance, the DNS-HS test is designed to assess the coordinated function of core stability. Hence, it may be more appropriate for individuals with acute LBP and associated core instability. However, the validity and reliability of these core stability testing methods have not been demonstrated. The present study intended to establish criterion validity and intra-rater test-retest reliability of the DNS-HS test by comparing conventional BSLLT measurements using Simi kinematic and ultrasound measurements. We hypothesized that the DNS-HS test would be a valid and reliable measure of core stability in individuals with core instability.

2. Methods

2.1 Participants

Twenty young adults with core instability (18 men, 19 women) participated in the experimental test, which was approved by a major university Human Studies Committee (2013-03). All participants provided informed consent prior to participation. Those participants who were unable to perform the BSLLT while maintaining target pressure level (40 $\pm$ 10 mmHg) recorded from PBU with no compensation were included [5, 6, 7]. Exclusion criteria were any known musculoskeletal impairments, surgery, pain or injury to the low back or lower extremities, or medications that may affect experimental tests.

2.2 Experimental procedure

All subjects underwent a physical health screening and BSLLT to check for core instability [5, 6, 7]. A procedural checklist and standardized verbal instructions were followed to ensure consistent experimental procedures. All measurement systems were calibrated for each participant before data acquisition.

Ultrasound (The SonoAce 6000, Medison Co., Ltd, Korea) in M-mode with a 3.5 MHz curved transducer (HC2-5) was used to measure diaphragm movement. A pressure biofeedback unit (PBU, Chattanooga Group, Hixon, TN, USA) containing a 3-chamber pressure bag connected to a pressure gauge and inflation device was used to detect core instability [12]. 3D motion capture system (Simi Reality Motion Systems, SRMS, Simi Aktisys GmbH, Unterschleissheim, Germany) was used to determine the hip joint angle during core stability tests. The core stabilization tests included BSLLT and DNS-HS. BSLLT is considered a gold standard measurement against which the new DNS-HS measurement was validated (Fig. 2).

2.3 Criterion validity

Criterion validity was determined by comparing hip joint angle data that were sequentially assessed by BSLLT and DNS-HS methods. The hip joint angle during BSLLT was determined first. For BSLLT, a participant was initially positioned in the supine position with hip 90 ${}^{\circ}$ flexion and knee extension. He or she was asked to contract the abdominal muscles concentrically into the lumbar spine and to consciously maintain this stabilization position. Participants then gradually lowered both legs toward the floor without any lumbopelvic movement. For DNS-HS, participants were initially positioned in a hook-lying position with 70 ${}^{\circ}$ of hip flexion. He or she was asked to breathe in and subconsciously or automatically descend the diaphragm to activate the TrA, multifidus, and pelvic floor chain muscles. The investigator (JJL) anteriorly palpated the xiphoid process, laterally palpated 10 $\sim$ 12 ribs, and posteriorly palpated the angulus costae to ensure symmetrical activation while expanding the lower ribs (10 $\sim$ 12th ribs) in the lateral direction. The corrective movement involves caudal movement and widening of the intercostal spaces, and relatively stable rib motion (no cranial motion) in a transverse plane [13]. Once core stabilization was obtained, the participants were asked to breathe naturally and then slide both heels along the floor without any lumbopelvic movement.

Prior to data collection, all subjects underwent approximately 10 practical sessions 5 minutes per each session, 5 minutes rest) to become familiarized with both BSLLT and DNS-HS methods. Those participants who successfully performed the tests were included in data collection. A PBU was placed under the 5 ${}^{\text{th}}$ lumbar spine and inflated to 40 mmHg, and participants were then instructed to maintain the target pressure range (40 $\pm$ 10 mmHg) during BSLLT or DNS-HS tests. Proper diaphragm movement and abdominal muscle activation were concurrently monitored with an ultrasound with a 3.5 MHz curved transducer. The transducer was placed between midclavicular and anterior axillary lines in the subcostal area, and applied to the medial, cranial, and dorsal directions [14]. Diaphragm movement was measured by placing electronic calipers just inside two hyperechoic fascial lines outlining the diaphragm where two parallel lines were obtained at the resting end expiration conditions [15].

Hip joint angle data were obtained during each BSLLT and DNS-HS measurement. Specifically, a participant was instructed to lower or slide their legs toward the plinth while maintaining neutral lumbopelvic alignment. A test was terminated whenever the PBU pressure decreased, and the hip joint angle was then measured with the Simi Aktisys. Hip joint angle kinematics were determined by Simi motion analysis in which three passive markers were placed on the lateral abdominal wall at the L3 level, the greater trochanter, and the lateral femoral epicondyle. A Basler acA640 camera (Basler Ahrensburg, Germany) with a sampling rate of 60 Hz was set a perpendicular distance of 1 m away from the participant (Fig. 1). Three hip joint angle measurements during each BSLLT and DNS-HS test were averaged for further statistical analysis.

Figure 1.

DNS-HS test.

Figure 2.

Flow chart of validity and reliability of test.

2.4 Test-retest reliability

Test-retest reliability was established by determining intra-rater DNS-HS measurement consistency. The reliability test was performed on two separate occasions, approximately 24 hours apart. All testing conditions were maintained as consistently as possible, with the same investigators, experimental procedures (i.e., consistent instruction, calibration, testing sequence, established hook-lying positions), time of day, interval, and testing environment (lighting, temperature). All participants in this study underwent practice trials to get familiarized with the repeated test measurements. Data were only collected after participants showed understanding of the experimental tests. Independent sets of hip joint angle measurements from the same participants were recorded for data analysis.

2.5 Statistical design and analysis

SPSS Statistics ver. 21.0 software (SPSS, Inc., Chicago, IL, USA) was used for all statistical analyses. Descriptive statistics including means and standard deviations were computed. The criterion validity and intra-rater test-retest reliability were determined using intraclass correlation coefficients, (ICC ${}_{2,3}$ ) and (ICC ${}_{3,3}$ ) respectively [16]. Paired $t$ -tests were used to determine any significant difference in demographic characteristics between gender and mean hip joint angles for BSLLT measurements, DNS-HS measurements, and repeated DNS-HS measurements. A significance level of $P<$ 0.05 was set for all statistical analyses.

3. Results

Table 1 presents the demographic characteristics of participants. The validity of the DNS-HS test was determined by comparing hip joint angle data between conventional BSLLT and DNS-HS measurements. Criterion validity was found to be (ICC ${}_{2,3}$ ) $=$ 0.869 (95% CI), suggesting a good relationship between the two core stability measures. Table 2 shows results of the test-retest reliability of examiner repeated measurements of the DNS-HS method using Simi Reality Motion Systems. Test-retest reliability was found to be (ICC ${}_{3,3}$ ) $=$ 0.974 (95% CI), suggesting consistency between the repeated DNS-HS measurements.

Table 1
Represents demographic characteristics of asymptomatic participants who demonstrated core instability ( $N=$ 37)

Parameters
Gender (M/F)	18/19
Age	21.11 $\pm$ 2.36
Height	169.68 $\pm$ 7.48
Weight	63.97 $\pm$ 8.86
BMI	22.19 $\pm$ 2.40

Table 2

Validity obtained by comparing BSLLT and DNS-HS measurement data

	BSLLT	DNS-HS	ICC (2, k)	$P$ -value
Hip angle ( ${}^{\circ}$ )	74.27 $\pm$ 9.44 ${}^{*}$	47.48 $\pm$ 9.00	0.869	$<$ 0.001 ${}^{\dagger}$

${}^{*}$ Mean $\pm$ SD; Validity ICC was computed based on 95% CI with 0.749–0.933 range. ${}^{\dagger}$ Significant difference, $P<$ 0.05.

Additionally, mean hip angle data showed a significant difference between the BSLLT (74.27 $\pm$ 9.44) and DNS-HS (47.48 $\pm$ 9.00) measurements.

4. Discussion

The present study established criterion validity and test-retest reliability of a novel core stability test. There is a dearth of clinical evidence regarding the established validity and test-retest reliability of core stability measurements. Core instability is one of the most perplexing and clinically challenging neuromusculoskeletal conditions because multiple attributes involve core stabilization, strength, power, endurance, flexibility, motor control, and function. Furthermore, core instability lacks a universal definition. Currently, there is no gold standard assessment of core stability with established evidence for its validity and reliability. This makes it difficult to correlate the validity and reliability data of the DNS-HS test with any other core stability measurement.

Table 3
Test-retest reliability of DNS-HS measurement

	Test	Re-test	ICC (3, k)	$P$ -value
Hip angle ( ${}^{\circ}$ )	47.48 $\pm$ 9.00 ${}^{*}$	47.66 $\pm$ 9.20	0.974	$<$ 0.001 ${}^{{\dagger}}$

${}^{*}$ Mean $\pm$ SD; Reliability ICC was computed based on 95% CI with 0.949–0.986 range. ${}^{{\dagger}}$ Significant difference, $P<$ 0.05.

The present criterion validity study compared the DNS-HS core stability test with gold standard BSLLT and demonstrated a good relationship between the two measures (ICC $=$ 0.869). This finding indicates that the DNS-HS core stability test is a valid test for core stability in individuals with core instability. The BSLLT is a universally accepted clinical assessment of abdominal muscle strength that is presumed to characterize abdominal core stability [10, 11, 17, 18, 19, 20, 21].

Based on extensive literature review, only one validity study related to core stability exists to date [20]. Gilleard and Brown developed an abdominal muscle strength test (AMT) and first established its validity by correlating the AMT level of difficulty and electromyographical (EMG) activity in 22 subjects. EMG data collected from the upper and lower rectus abdominis and external and internal oblique muscles were used to validate with four AMT levels of difficulty. The EMG activity results showed increased muscle amplitudes from Levels 1 to 3 whereas at Level 4, the external and internal oblique muscles had relatively increased muscle amplitudes [20]. This indicates that the AMT is a valid measure of discerning abdominal muscle strength in Levels 1 to 3. A different abdominal muscle was utilized in Level 4, which better reflected oblique muscle strength, warranting a different muscle test measurement. However, the validity of AMT has not been fully determined yet. It remains unclear if the abdominal muscle strength recorded by AMT truly represents core stability.

Test-retest data in the present study revealed ICC $=$ 0.974, representing excellent consistency or repeatability in DNS-HS core stability measurements. This finding was consistent with pervious intra-session measurements of BSLLT [17]. Zannotti et al. [17] examined the kinematics of the double-leg-lowering (DLL) test (also known as ‘BSLLT’) of abdominal muscle strength in healthy young adults using ICC for intra-session measurements of pelvic tilting occurring at 5 hip joint angles (15 ${}^{\circ}$ , 30 ${}^{\circ}$ , 45 ${}^{\circ}$ , 60 ${}^{\circ}$ , and 75 ${}^{\circ}$ ) during the DLL test under controlled or uncontrolled pelvic stability conditions [17]. Intra-session ICC measures recorded early in the DLL test were less than acceptable, although ICCs for the latter hip angles were more acceptable and higher than the initial angles, ranging from 0.63–0.95 [17]. For the first (1.0 ${}^{\circ}$ ) pelvic tilting angle, ICCs were 0.55 for controlled pelvic stability and 0.24 for uncontrolled pelvic stability [17]. Furthermore, the DLL test presents a considerable challenge to the abdominal muscles. None of the tested young healthy subjects were able to successfully stabilize pelvic tilting, jeopardizing the internal validity of the scoring system associated with the DLL test.

Similarly, Smidt et al. [10] assessed the validity of the DLL, sit-up, and prone trunk extension clinical tests in normal adults and adults with a history of LBP. Although the DLL test was superior to other tests, only global abdominal strength was estimated, not core stability function. The majority of subjects were unable to complete the Grade III-IV levels test. Hence, these tests might not be appropriate for individuals with acute or subacute LBP. For these reasons, the DLL test has been challenged as a valid or reliable measure of core instability that is applicable for most clinical populations with LBP [10, 11, 17, 18, 22].

Additionally, mean hip angle data differed between the BSLLT (74.27 ${}^{\circ}$ $\pm$ 9.44 ${}^{\circ}$ ) and DNS-HS (47.48 ${}^{\circ}$ $\pm$ 9.00 ${}^{\circ}$ ) measurements because initial hip joint angle and level of difficulty between the two measurements were dissimilar. For example, initial hip joint flexion with BSLLT was 90 ${}^{\circ}$ whereas the DNS-HS angle was 70 ${}^{\circ}$ . This initial angle was unlikely to affect the results. However, the difficulty level between the two core stability measurements may have influenced resulting mean hip angle data. The DNS-HS was relatively easier to perform than the BSSLT because DNS-HS provided a better graded counterbalancing (hip flexor) moment to maintain a neutral pelvic position. For example, lumbar hyperextension and anterior pelvic tilt movements were commonly detected during BSSLT. This is due to variation in the size of the gluteal mass precluding standard positioning of the pelvis or relative distensibility (shortness) of the hip flexors, respectively [17]. However, the lack of standardization and inadequate range of difficulty associated with BSSLT was not evident with the newly developed DNS-HS test. Moreover, the DNS-HS test was more sensitive and discriminative at detecting minute differences in core stability. During the BSLLT, average hip (extension) angle was approximately 10 ${}^{\circ}$ (from initial hip angle 90 ${}^{\circ}$ to 80 ${}^{\circ}$ ) of a 0–90 ${}^{\circ}$ range (12%). During the DNS-HS test, average hip (extension) angle was approximately 20 ${}^{\circ}$ (29%) (Table 3).

Taken together, our collective findings suggest that the DNS-HS measurement may be useful to differentiate original source(s) of spinal pathology resulting from either pain mediated inhibition of core muscles or core instability itself because it may be more conducive and sensitive to individuals with acute or severe low back pain associated with core instability. Nevertheless, this hypothesis should be validated in future research in individuals with acute or subacute low back pain.

Several research limitations should be taken into consideration for future studies. First, the present study included individuals with core instability and a previous history of low back pain. Hence, it is difficult to generalize our results to pathological populations with acute or subacute LBP. Second, the initial starting position differed for the BSLLT (90 ${}^{\circ}$ hip flexion in supine) and bilateral DNS-HS (70 ${}^{\circ}$ in a hook-lying position) tests, which may have confounded the results. However, the bilateral heel sliding movement employed in the DNS-HS test is a preferred exercise for individuals with LBP compared to the bilateral double leg lowering movement [7]. Lastly, we examined hip kinematics between the BSLLT and DNS-HS tests and found high correlations because kinematic measurement was consistent. It is of great interest to measure EMG activity and muscle thickness to determine motor control issues associated with the level of difficulty during core stability. Nevertheless, the functional relationship between abdominal muscle tests and EMG activities showed an inconsistent and low correlation ( $r=$ 0.18) [20].

5. Conclusions

The present study established the validity and excellent test-retest reliability of a novel DNS-HS core stability test. Clinically, the DNS-HS test is useful to objectively quantify core stability and aid in early detection. It is also useful for evaluation and monitoring of even minute progress after spinal rehabilitation in individuals with low back pain and core instability.

Conflict of interest

None to report.

Footnotes

Acknowledgments

This research was in part supported by a Brain Korea 21 PLUS Project grant (NO. 2016-51-0009) of Korean Research Foundation awarded to the Department of Physical Therapy of the Graduate School, Yonsei University.

References

Frank

Kobesova

Kolar

. Dynamic neuromuscular stabilization & sports rehabilitation. Int J Sports Phys Ther 2013; 8(1): 62-73.

Hides

Stanton

Freke

Wilson

McMahon

Richardson

. MRI study of the size, symmetry and function of the trunk muscles among elite cricketers with and without low back pain. Br J Sports Med 2008; 42(10): 809-813.

Richardson

Hodges

Hides

. Therapeutic exercise for lumbopelvic stabilization: a motor control approach for the treatment and prevention of low back pain. Churchill Livingstone. 2004.

McGill

Childs

Liebenson

. Endurance times for low back stabilization exercises: clinical targets for testing and training from a normal database. Arch Phys Med Rehabil 1999; 80(8): 941-944.

Faries

Greenwood

. Core training: Stabilizing the confusion. J Strength Cond Res 2007; 29(2): 10-25.

Ferreira

Hodges

. Changes in recruitment of the abdominal muscles in people with low back pain: ultrasound measurement of muscle activity. Spine (Phila Pa 1976) 2004; 29(22): 2560-2566.

Stanton

Reaburn

Humphries

. The effect of short-term Swiss ball training on core stability and running economy. J Strength Cond Res 2004; 18(3): 522-528.

Evans

Refshauge

Adams

. Trunk muscle endurance tests: reliability, and gender differences in athletes. J Sci Med Sport 2007; 10(6): 447-455.

Shields

and Heiss

. An electromyographic comparison of abdominal muscle synergies during curl and double straight leg lowering exercises with control of the pelvic position. Spine (Phila Pa 1976) 1997; 22(16): 1873-1879.

10.

Smidt

Blanpied

. Analysis of strength tests and resistive exercises commonly used for low-back disorders. Spine (Phila Pa 1976) 1987; 12(10): 1025-1034.

11.

Smidt

Blanpied

Anderson

White

. Comparison of Clinical and Objective Methods of Assessing Trunk Muscle Strength-An Experimental Approach. Spine (Phila Pa 1976) 1987; 12(10): 1020-1024.

12.

von Garnier

Köveker

Rackwitz

Kober

Wilke

Ewert

Stucki

. Reliability of a test measuring transversus abdominis muscle recruitment with a pressure biofeedback unit. Physiother 2009; 95(1): 8-14. doi: 10.1016/j.physio.2008.10.003.

13.

Liebenson

. Rehabilitation of the spine: a practitioner’s manual. Lippincott Williams & Wilkins. 2007.

14.

Boussuges

Gole

and Blanc

. Diaphragmatic motion studied by m-mode ultrasonography: methods, reproducibility, and normal values. Chest 2009; 135(2): 391-400. doi: 10.1378/chest.08-1541.

15.

Ayoub

Cohendy

Dauzat

Targhetta

De la Coussaye

Bourgeois

Ramonatxo

Prefaut

Pourcelot

. Non-invasive quantification of diaphragm kinetics using m-mode sonography. Can J Anaesth 1997; 44(7): 739-744.

16.

Portney

Watkins

. Foundations of clinical research: applications to practice. Vol. 2. Upper Saddle River, NJ; Prentice Hall. 2000.

17.

Zannotti

Bohannon

Tiberio

Dewberry

Murray

. Kinematics of the double-leg-lowering test for abdominal muscle strength. J Orthop Sports Phys Ther 2002; 32(9): 432-436.

18.

Cutter

Kevorkian

. Handbook of manual muscle testing. McGraw-Hill/Appleton & Lange. 1999.

19.

Kendall

McCreary

Provance

. Muscle testing and function. Philadelphia: Lippincott Williams & Wilkins. 1993.

20.

Gilleard

Brown

. An electromyographic validation of an abdominal muscle test. Arch Phys Med Rehabil 1994; 75(9): 1002-1007.

21.

Moreland

Finch

Startford

Balsor

Gill

. Interrater reliability of six tests of trunk muscle function and endurance. J Orthop Sports Phys Ther 1997; 26(4): 200-208.

22.

Smith

Rasmussen

Lechner

Gossman

Quintana

Grubbs

. The effects of lumbosacral support belts and abdominal muscle strength on functional lifting ability in healthy women. Spine (Phila Pa 1976) 1996; 21(3): 356-366.