Differences in physical examination findings between those who present with or without headache soon after a whiplash injury: a cross-sectional study

ABSTRACT

Aim

To determine differences in physical examination findings between people with acute whiplash-associated disorders (WAD) with and without headache.

Methods

In this cross-sectional study, participants with acute WAD were evaluated to assess differences in the presence of physical impairments. The following were assessed: pain intensity on manual palpation the over spinous process of C1-C3, zygapophyseal joints of C0-C4, and trapezius, sternocleidomastoid, suboccipitalis, masseter and temporalis muscles; cervical range of motion (ROM); flexion-rotation test (FRT); forward head posture; cranio-cervical flexion test (CCFT); neck flexor and extensor endurance; pressure-pain thresholds (PPT) over neural structures and upper limb neural tests (ULNT) in addition to median UNLT + CCF. Correlation analyses were performed to assess the association between examination findings and headache intensity. Logistic regression and discriminant analyses were also performed.

Results

Forty-seven participants (26 men and 21 women; mean age = 38.9 years old) were included in the study. 60% of the participants presented with headache. Several examination findings were significantly different between groups. A group of examination findings composed of neck endurance, manual palpation over cervical and muscular structures, PPT, CCFT, ROM and FRT could discriminate between groups with a sensitivity of 86.7% and specificity of 90%.

Conclusions

Several neuromusculoskeletal features are different between people with acute WAD with or without headache. A combination of features could distinguish between groups with high levels of sensitivity and specificity.

Introduction

Headache is one of the most common symptoms after a whiplash injury, being present in up to 57% of people with acute whiplash-associated disorders (WAD) [1,2]. According to the International Headache Society (IHS), acute headache attributed to a whiplash injury must occur, or increase if headache was already present, within 7 days after the injury [3].

Following a motor vehicle accident, injury to the cervical region can contribute to the sensation of pain in the head [4,5]; based on the convergence of sensory neurons on the trigemino-cervical nucleus, nociceptive inputs from cervical afferents, mainly upper cervical afferents, can elicit pain in the craniofacial region [6–9]. In particular, pain sensations from segments C3-C4 upwards can refer pain to the head [10]. These peripheral inputs may contribute to hyperexcitability in the trigemino-cervical complex (TCC), eliciting or contributing to the presence of headache [7]. Previous work has demonstrated that hyperexcitability in the TCC can be present in people with chronic WAD [11].

Several studies have described the presence of cervical musculoskeletal impairments in different headache conditions and have suggested that they may play a role in the presence of headache [12,13]. Additionally, in people with WAD, various neuromusculoskeletal and sensory impairments have been identified, such as altered range of motion, the presence of trigger points, reduced neck strength or mechanosensitivity of neural structures [14–17]. Nevertheless, in relation to headache, no studies have examined such impairments to understand potential differences between people with WAD with (WAD H) or without headache (WAD NH). Previously, a number of neuromusculoskeletal features have been described in people with headache conditions, including patients with migraine or tension-type headache, when compared to healthy controls [18,19], and also between patients with cervicogenic headache versus migraine [20]. Although people with WAD will likely present with neuromusculoskeletal impairments regardless of the presence of headache, it may be that some features differ between those who present with headache and those who do not, since some of these features may contribute to the presence of headache. In addition, an understanding of the association between the extent of physical examination findings and headache intensity may offer further understanding of the role of physical features in the presentation of headache following a whiplash injury.

Therefore, considering the potential role of cervical neuromusculoskeletal features in the presentation of headache after a whiplash injury, the current study aimed to 1) assess differences in physical findings between people with acute WAD who present with or without headache; 2) determine whether the extent of neuromusculoskeletal impairments are correlated with headache intensity; and 3) evaluate if a group of physical examination tests can discriminate between people with WAD H and WAD NH. We hypothesized that cervical neuromusculoskeletal impairments would be more present in those with acute WAD presenting with headache.

Methods

Study design

A case-control study was conducted on people with acute pain attributed to a whiplash injury and was performed at Clínica San Vicente, Madrid, Spain. Data were collected between September 2020 to February 2021. Ethical approval was granted by the institutional human research ethics committee from University Rey Juan Carlos, Madrid, Spain (Ref: 1003202108121). The study adhered to the Declaration of Helsinki and is reported in accordance with STROBE guidelines [21,22].

Participants

People with a diagnosis of WAD attributed to a motor vehicle accident in the last 30 days were recruited consecutively. In order to have a more homogenous group and considering that Grade II represents the majority of patients with WAD [23], we only included those with WAD grade II according to the Quebec Task Force [24]. Additionally, we wanted to minimize possible outliers in results which may have occurred if people with grades III-IV performed some of physical tests due to increased disability and pathology. According to the IHS [3], people with a previous headache were included in the headache group if their headache intensity increased within the 7 days after the accident. No participants were recruited before 7 days after the accident to respect this criterion.

Individuals were excluded if they were diagnosed with fibromyalgia or had a history of generalized pain, had experienced a previous whiplash injury, had been diagnosed with osteoporosis, cervical myelopathy, had a diagnosed temporomandibular disorder, vertebral fractures and/or, inflammatory or rheumatic diseases, had a known psychological disorder, congenital disturbances, had undergone previous surgery in the cervical region, had received physical therapy treatment after the accident but before participation in the study, or were not able to complete patient-reported outcome measures. In addition, with the aim of excluding people suffering from concussion, we followed the criteria of the IHS [3] and excluded people that had experienced one or more of the following signs and/or symptoms: confusion, disorientation or impaired consciousness; loss of memory for events immediately before or after the accident; and one or more of the following: nausea, vomiting, visual disturbances, vertigo, gait and/or postural imbalance, and impaired memory and/or concentration.

After being diagnosed by a physician, who then informed the participants about the study, those that agreed to participate provided written informed consent and were referred to the Physiotherapy Department. Patients with a previous headache condition which increased after the accident, were asked to state their previous headache diagnosis before referral to the Physiotherapy Department. Participants were advised not to disclose their headache status when attending the Physiotherapy Department for assessment and therefore the assessor was blinded to group allocation. For all participants, if the previous headache condition was present on 15 or more days per month, they were considered to have chronic rather than episodic headache [3].

The sample size estimation was performed using the Grammo calculator v.7.12. Due to the number of physical tests included in our study, we chose to base our sample size calculation on the pressure-pain thresholds (PPT), given that this is one of the most commonly used physical tests reported in the literature in people with WAD and hypersensitivity may be related to a poor recovery, suggesting its importance [17]. Based on detecting a significant between-group differences of 20% for PPT, with an alpha level of 0.05 and a desired power of 80%, a sample size of at least 16 participants per group was required [11].

Procedures

Before the evaluation, participants were advised that these physical tests would not be considered for their final health report, and therefore would not be reviewed as part of any insurance claim.

All measurements were collected in a single session conducted within the Physiotherapy Department by the same rater, who is a physical therapist with four years of experience at the time of testing and with a Master’s Degree in Orthopaedic Manual Therapy.

All measures were evaluated twice, and the mean of both was used for the analysis. When tests were performed bilaterally, the outcomes were considered separately for the most and least painful side (determined based on self-report). This decision was made on the basis that considering the mean of both sides may affect the results since many patients could only present with unilateral impairments.

Outcome measures

Age, sex, height, and weight were recorded for all participants. A short explanation is presented in this section and a further explanation of the evaluation procedures for each measurement can be found in Appendix 1.

Articular system

Cervical Range of Motion (CROM,º)

Flexion, extension, lateral-flexion and rotation were assessed in a relaxed sitting position using a smartphone Xiaomi® MiA2 with Smartphone apps [13,25,26]. Intra-rater reliability ranges from 0.489 to 0.917 [27].

Passive Accessory Intervertebral Movements (PAIVMs)

Central and bilateral posterior-anterior intervertebral movements were applied as a grade III over C1-C3 (central, spinous processes) and C0-C1/C3-C4 (bilateral, zygapophyseal joints). Since many physical tests were conducted, and to avoid an increase in sensitivity due to extended testing we chose only to assess C0-C4 segments. This was also justified considering the main role of upper cervical structures in the pathophysiology of headache. Pain intensity provoked through movement was recorded on a Numerical Rating Scale (NRS), ranging from 0 (no pain) to 10 (worst pain imaginable) [28,29]. The intra-rater reliability of these tests ranges from 0.885 to 0.977 [27].

Flexion-Rotation Test (FRT)

The participant lay in supine on the plinth. They were asked to relax while their neck was passively moved to end range cervical flexion by the examiner. In this flexed position, the head and neck were passively rotated as far as possible within comfortable limits, and the number of degrees (º) of rotation was recorded with the Smartphone Compass Application. The test was performed bilaterally [30]. The intra-rater reliability of this test ranges from 0.896 to 0.936 [27].

Forward Head Posture (FHP)

FHP was assessed in a relaxed standing and sitting position via a lateral photograph taken from a distance of 1.5 meters and the FHP app was used to calculate the Cranio-Vertebral angle [31,32]. The intra-rater reliability of this test ranges from 0.859 to 0.937 [27].

Muscular system

Muscle palpation

Palpation was performed at predetermined points over different muscles. The upper trapezius, suboccipitalis, masseter, temporalis and sternocleidomastoid (SCM) were assessed bilaterally to measure the sensitivity to pressure in muscles which have previously been described as referring pain to the head. Pain intensity was recorded via an NRS, ranging from 0 (no pain) to 10 (the worst pain imaginable). The intra-rater reliability of these tests ranges from 0.937 to 0.990 [27].

Cranio-Cervical Flexion Test (CCFT)

The participant lay supine with the neck in a neutral position, supported by towels under their head as needed. An uninflated pressure cuff (Chattanooga Stabilizer Group Inc., Hixson, TN, USA) placed behind the upper cervical region so that it abutted the occiput and was then inflated to a baseline pressure of 20 mmHg. The participant then performed the stages of the CCFT and the highest level of the test held for 10 seconds was recorded, as described previously [29,33]. The intra-rater reliability of this test is ICC = 0.920 [27].

Neck flexor endurance

The test was performed with the participant positioned in supine on the plinth. The participant’s head was positioned in slight upper cervical flexion by the examiner who placed his left hand on the table just below the participant’s occiput and seconds holding this position were measured [34]. Concerning intra-rater reliability, an ICC (95%CI) = 0.933 (0.878–0.963) has been reported [27].

Neck extensor endurance

This test measured the time, in seconds, that the participant could keep their head steady, while lying in prone with the head over the edge of the plinth in a neutral position [35]. Concerning intra-rater reliability, an ICC (95%CI) = 0.962 (0.933–0.979) has been reported [27].

Neural system

Mechanosensitivity of the median, radial and ulnar nerves

Upper limb neurodynamic tests (ULNT) for the median, radial and ulnar nerves were assessed as described previously [36]. The degrees (º) of elbow movement was recorded. The intra-rater reliability of this test ranges from 0.310 to 0.977 [27].

Mechanosensitivity during Upper Limb Neurodynamic Testing for the median nerve combined with Cranio-Cervical Flexion (CCF)

The patient was asked to perform active CCF and then the ULNT1 was performed as described previously [37]. The degrees (º) of elbow movement was recorded. The intra-rater reliability of this test ranges from 0.734 to 0.0.904 [27]

Pain Pressure Thresholds over the median, radial, ulnar, supra-orbital and greater occipital nerves

Pressure pain thresholds were measured bilaterally using a digital algometer (Force Ten^TM-Model FDX, Wagner, Greenwich, USA) with a surface area of round tip of 1 cm² and were recorded in N/cm². The assessment of all the nerve trunks were performed following the procedure reported in previous studies [38,39]. Intra-rater reliability ranges from 0.893 to 0.977 [27].

Tests were performed twice, in two sets, with ten minutes of rest between repeated testing for the assessment of each set. During this time, participants sat on a chair resting. The order of testing was comparable between sets and endurance/motor control tests were always performed at the end of each set to avoid the possible influence of hypoalgesic effects of exercise. The order of testing was chosen to minimize change in the patients position and adhered to the following sequence: CROM, FHP, PAIVMs, PPT over the GON, FRT, muscle palpation, PPTs over the other nerves, ULNTs and ULNT + CCF, CCFT, neck flexor endurance, neck extensor endurance.

Statistical analysis

All analyses were performed using SPSS 25.0 (IBM Corp., Armonk, NY, USA). Descriptive statistics were generated and between group (headache versus no headache) differences were examined for participant demographics, patient reported outcome measures and physical examination findings. The normality of data distribution was assessed using the Shapiro-Wilk’s Test. Based on the normality test, differences between groups were assessed using the Student’s T-Test for parametric data, and Mann-Whitney U test for non-parametric data. X2 was performed for categorical data.

Bivariate correlations between headache intensity and physical examination findings were performed. Pearson’s correlation coefficient was used if the data were normally distributed, and Spearman’s correlation was used if data followed a non-normal distribution. A weak correlation was considered when the correlation was −0.3 to 0.3, except for 0, which meant no correlation. Moderate correlations were considered when the correlation was −0.3 to −0.5 and 0.3 to 0.5 and values ranging from −0.5 to −0.7 and from 0.5 to 0.7 were considered to be a strong correlation. Higher values, up to −1 and 1, respectively, were considered to be a very strong correlation [40]

A discriminant analysis was used to investigate whether a pattern of physical impairment distinguished those with and without headache [41]. To reduce the number of variables included in our model, as a first step, we collapsed variables. Cervical ROM was averaged across all movement directions and collapsed into one variable, which represents average cervical ROM. The same procedure was followed for endurance (END), PPT, posture (POST), ULNT, pain intensity through manual palpation of the cervical spine (NRSC) and cervical muscles (NRSM), CCFT and FRT. Sensitivity and specificity were also calculated [38]. Variables which did not demonstrated statistically significant differences between groups were excluded from the discriminant analysis. The Canonical correlation was calculated to evaluate the association between the discriminant scores and the groups (a value close to 1 show that the function discriminates well). To assess the proportion of the total variance in the discriminant scores not explained by differences among groups, Wilk’s Lambda was calculated (values near 0 indicate the group’s mean Discriminant scores differ); Chi-Square and its significance were reported to indicate whether a highly significant difference between the group’s centroids existed [41]. The Canonical correlation was calculated to report the association between the discriminant scores and the groups, with a high value (near 1) showing that the function discriminates well [41]. Sensitivity and specificity were also calculated, and a cross-validation was performed according to the test-retest method [41] taking approximately 70% of the cases as the test sample. For all analyses, significance was set to p < 0.05.

Results

Forty-nine people were assessed and after the exclusion of two with a history of neck surgery, 47 patients remained in the study. Among them, 28 participants (59.6%; 16 women) presented with headache. Five of these patients suffered from previous headaches which had increased after the whiplash injury: 2 presented with migraine (1 episodic and 1 chronic) and 3 tension-type headache (1 episodic and 2 chronic). Nineteen (40.2%; 5 women) were considered as controls due to the absence of headache. No significant differences between groups were found for age, sex, height, weight or days from the accident to the assessment (Table 1).

Table 1.

Descriptive statistics. Data expressed on mean (SD)(n = 47).

Variables	Group		Z
	Headache(n = 28)	No Headache(n = 19)	P
Age (years)a	37.6 (11.1)	40.9 (10.9)	0.319
Genderb(male/female)	12/16	14/5	0.064
Height (cm)a	174.5 (8.8)	177.1 (9.9)	0.370
Weight (kg)a	70.7 (10.1)	76.6 (10.4)	0.064
Daysa	13.4 (4.3)	11.7 (3.7)	0.152
VAS Headachea (mm)	47.4 (14.2)	–	–

^aT-Student; ^bChi-Square

VAS: Visual Analogue Scale

Summary statistics and between group differences for all measurements can be found in Table 2. Compared to WAD NH, the highest significant differences (p < 0.001) were found for pain intensity on palpation of C1 (Mean Difference [MD] = 1.6) and C2 (MD = 1.7), C0-C1 (MD = 1.7) and C1-C2 (MD = 1.6) on the least painful side, and C1-C2 on the most painful side (MD = 2.2); ROM during the FRT to most (MD = 7.4º) and least (MD = 5.7º) restricted sides; pain intensity on palpation of the temporalis muscle on the most painful side (MD = 1.8); CCFT performance (MD = 1.8); endurance of the flexors (MD = 4.2 s) and extensors (MD = 5.7 s); and PPT over least (MD = 3.7 N/cm²) and most (MD = 3.5 N/cm²) sides of the GON.

Table 2.

Between group differences in physical examination findings between people with acute WAD with and without headache. Mean (SD).

Test	Headache Group (n = 28)	Non-Headache Group (n = 19)	P-value
C1 (NRS)	6.00 (1.6)	4.4 (1.6)	<0.001a,*
C2 (NRS)	6.3 (1.2)	4.6 (1.3)	<0.001b,*
C3 (NRS)	5.7 (1.6)	5.00 (1.50)	0.078a
Most painful C0-C1 (NRS)	6.2 (1.5)	4.4 (2.0)	0.001b,*
Least painful C0-C1 (NRS)	4.9 (1.5)	3.2 (1.7)	<0.001a,*
Most painful C1-C2 (NRS)	6.8 (1.1)	4.6 (1.5)	<0.001a,*
Least painful C1-C2 (NRS)	5.4 (1.1)	3.8 (1.2)	<0.001a,*
Most painful C2-C3 (NRS)	6.3 (1.2)	5.21 (1.3)	0.007b,*
Least painful C2-C3 (NRS)	5.3 (1.5)	4.1 (1.4)	0.005a,*
Most painful C3-C4 (NRS)	6.1 (1.0)	5.6 (1.1)	0.057a
Most painful C3-C4 (NRS)	5.1 (1.2)	4.4 (1.7)	0.202b,
FLEX (º)	51.4 (13.9)	60.2 (10.3)	0.012a,*
EXT (º)	31.2 (5.8)	34.2 (6.5)	0.052a
Most restricted LF (º)	29.3 (6.6)	32.8 (4.7)	0.026a,*
Least restricted LF (º)	33.4 (6.5)	36.9 (4.4)	0.021a,*
Most restricted ROT (º)	48.3 (13.4)	58.9 (12.9)	0.005a,*
Least restricted ROT (º)	54.5 (13.2)	65.2 (11.7)	0.003a,*
Most restricted FRT (º)	26.6 (6.3)	34.0 (4.3)	<0.001a,*
Least restricted FRT (º)	32.5 (6.9)	38.2 (3.8)	<0.001b,*
FHPSTAND (º)	50.5 (5.5)	51.7 (4.7)	0.218a
FHPSIT (º)	46.7 (3.3)	48.8 (4.8)	0.126b
Most painful SCM (NRS)	6.0 (1.5)	4.7 (1.5)	0.003a,*
Least painful SCM (NRS)	4.5 (1.7)	3.6 (1.6)	0.014a,*
Most painful TRAP (NRS)	6.9 (1.2)	5.6 (1.9)	0.004a, *
Least painful TRAP (NRS)	4.8 (1.4)	3.7 (1.4)	0.005a, *
Most painful SO (NRS)	6.3 (1.8)	4.8 (2.0)	0.015b,*
Least painful SO (NRS)	4.8 (1.8)	3.6 (1.8)	0.018a,*
Most painful MAS (NRS)	5.6 (1.3)	4.5 (1.5)	0.004a, *
Least painful MAS (NRS)	4.5 (1.4)	3.6 (1.3)	0.014b,*
Most painful TEMP (NRS)	5.9 (1.5)	4.1 (1.4)	<0.001b,*
Least painful TEMP (NRS)	4.5 (1.5)	3.2 (1.2)	0.004b,*
CCFT (22–30 mmHg)	24.1 (1.7)	25.9 (1.6)	<0.001b,*
FLEX MUSC (s)	10.2 (3.8)	14.4 (4.8)	<0.001a,*
EXT MUSC (s)	12.7 (5.7)	18.2 (4.8)	<0.001a,*
Least painful MEDN (kg/cm2)	18.7 (5.0)	21.7 (6.2)	0.040b*
Most painful MEDN (kg/cm2)	16.2 (3.5)	19.0 (5.6)	0.019a*
Least painful RADN (kg/cm2)	22.0 (7.3)	26.8 (6.6)	0.015a, *
Most painful RADN (kg/cm2)	20.7 (6.9)	22.4 (4.7)	0.024a*
Least painful ULNN (kg/cm2)	18.7 (5.6)	22.6 (6.4)	0.015b,*
Most painful ULNN (kg/cm2)	16.2 (4.6)	18.7 (5.00)	0.042a,*
Least painful SUPORBN (kg/cm2)	9.2 (3.3)	11.6 (2.2)	0.002b,*
Most painful SUPORB (kg/cm2)	8.1 (2.2)	10.1 (2.2)	0.001a. *
Least painful GON (kg/cm2)	9.6 (2.8)	13.3 (3.7)	<0.001a*
Most painful GON (kg/cm2)	8.0 (2.2)	11.5 (3.5)	<0.001a, *
Least restricted ULNT1 (º)	147.0 (11.6)	150.7 (9.8)	0.130a
Most restricted ULNT1 (º)	137.6 (9.8)	142.1 (11.6)	0.081a
Least restricted ULNT2(º)	31.9 (3.5)	32.6 (1.7)	0.198b
Most restricted ULNT2 (º)	29.1 (3.6)	30.1 (2.4)	0.139a
Least restricted ULNT3 (º)	120.8 (20.7)	110.5 (21.1)	0.052a
Most restricted ULNT3 (º)	106.9 (21.4)	92.5 (17.8)	0.010a*
Least restricted ULNT1+FCC (º)	143.9 (11.6)	148.2 (10.4)	0.102a
Most restricted ULNT1+FCC (º)	135.9 (10.3)	138.5 (11.1)	0.205a

^aT-Student; ^bU-Mann Whitney; * Statistical significance between groups (p < 0.05).

LF (Lateral-Flexion), Rot (Rotation), FRT (Flexion-Rotation Test), SUPORB (Supraorbitaire nerve), GON (Greater Occipital Nerve), PPT (Pressure Pain-Threshold), ULNT1 (Upper Limb Neural Test for Median), ULNT2 (Upper Limb Neural Test for Radial), ULNT3 (Upper Limb Neural Test for Ulnar), CCF (Cranio-Cervical Flexion), CCFT (Cranio-Cervical Flexion Test, SCM (Sternocleidomastoid), Trap (Trapezius), SO (Suboccipitalis), MAS (Masseter) TEMP (Temporalis); MEDN: Median nerve; RADN: Radial nerve; ULNN: Ulnar nerve.

Table 3 presents the correlations between physical examination findings and headache intensity. Strong correlations were found for pain intensity produced through manual palpation of C1-C2 (r = 0.622) and of the trapezius (r = 0.507) on the most painful side, FRT to the most restricted side (r = 0–.536), and for the PPT over the supraorbital nerve on the most painful side (r = −0.525) and over the GON on the most (r = −0.595) and the least (r = −0.577) painful sides.

Table 3.

Correlation between physical examination findings and headache intensity.

Most restricted side or non-side depending test	Correlation value	Least restricted side	Correlation value
C1a	0.496**	-	-
C2b	0.425**	-	-
C3a	0.218	-	-
C0C1b	0.458**	C0C1a	0.471**
C1C2a	0.622**	C1C2a	0.492**
C2C3b	0.225	C2C3a	0.246
C3C4a	0.146	C3C4b	0.090
FLEXa	−0.352*	-	-
EXTa	−0.292*	-	-
LFa	−0.328*	LFa	−0.358*
ROTa	−0.339**	ROTa	−0.351**
FRTRa	−0.536**	FRTb	−0.301*
FHPSTANDa	−0.018	-	-
FHPSITb	−0.222	-	-
SCMa	0.358*	SCMa	0.232
TRAPa	0.269	TRAPa	0.395**
SOb	0.333*	SOa	0.281
MASa	0.319*	MASa	0.297*
TEMPb	0.507**	TEMPb	0.430**
CCFTb	−0.418**	-	-
FLEXORSa	−0.448**	-	-
EXTENSORSa	−0.478**	-	-
MEDNb	−0.236	MEDNa	−0.282
RADNa	−0.337*	RADNa	−0.363*
ULNNb	−0.223	ULNNa	−0.339*
SUPORBNb	−0.525**	SUPORBNa	−0.444**
GONa	−0.595**	GONa	−0.577**
ULNT1a	−0.279	ULNT1a	−0.284
ULNT2a	−0.175	ULNT2a	−0.198
ULNT3a	0.396**	ULNT3a	0.277
ULNT1+FCCa	−0.216	ULNT1+FCCa	−0.272

^aPearson correlation coefficient; ^bSpearman’s correlation coefficient

Significant correlation was indicated in bold (p < 0.05(*) or p < 0.001 (**)). The absence or presence of a negative value (-) indicates a positive or negative correlation, respectively, indicating the direction of the correlation.

Abbreviations: LF (Lateral-Flexion), Rot (Rotation), FRT (Flexion-Rotation Test), SUPORBN (Supraorbitaire nerve), GON (Greater Occipital Nerve), PPT (Pressure Pain-Threshold), ULNT1 (Upper Limb Neural Test for Median), ULNT2 (Upper Limb Neural Test for Radial), ULNT3 (Upper Limb Neural Test for Ulnar), CCF (Cranio-Cervical Flexion), CCFT (Cranio-Cervical Flexion Test, SCM (Sternocleidomastoid), Trap (Trapezius), SO (Suboccipitalis), MAS (Masseter) TEMP (Temporalis) MEDN: Median nerve; RADN: Radial nerve; ULNN: Ulnar nerve.

Summary statistics and between groups differences for collapsed variables (END, PPT, POST, ULNT, NRSC, NRSM, CCFT and FRT), which were then used for the discriminant analysis, can be found on Appendix 2. Only collapsed variables which demonstrated significant differences between groups were included in the discriminant analysis and their discriminant coefficients were used to test the sensitivity and specificity of the group of physical measures to classify those who presented with headache from those who did not (Table 4). The eigenvalue of the canonical correlation was 0.72, demonstrating that the function discriminates well. Chi-Square test indicated that there was a highly significant difference between the groups’ centroids (X² = 30.04; p < 0.001). The Wilk’s Lambda showed that 45.8% of the total variance in the discriminant scores were not explained by differences among groups. Table 5 reports the sensitivity and specificity data; an overall sensitivity of 86.7% and a specificity of 90.00% were observed for the cross-validated procedure.

Table 4.

Discriminant function coefficients (standardized coefficients).

Test	Discriminant function
FRT	−0.005
PPT	−0.382
NRSC	0.554
NRSM	0.387
CCFT	−0.209
ROM	0.178
END	−0.191

FRT: Flexion-Rotation Test; PPT: Pressure-Pain Threshold; NRSC: Numerical Rating Scale for cervical structures; NRSM: Numerical Rating Scale for cervical muscles; CCFT (Cranio-Cervical Flexion Test), ROM (Range of Motion) and END: endurance.

The absence or presence of a negative value (-) indicates the likelihood of belonging to the headache group with higher values of each variable.

Table 5.

The sensitivity and specificity of physical measures to categorize those who presented with versus without headache.

			Predicted membership
		Group	No Headache	Headache	Total
Original classification	Count	No Headache	17	2	19
		Headache	3	25	28
	Sensitivity	No Headache	89.5%	10.5%	100.0%
	Specificity	Headache	10.7%	89.3%	100.0%
Cross-validated	Count	No Headache	13	2	15
		Headache	2	18	20
	Sensitivity	No Headache	86.7%	13.3%	100.0%
	Specificity	Headache	10.0%	90.00%	100.0%

Discussion

This case-control study examined differences in neuromusculoskeletal impairments and pain sensitivity between patients with and without headache shortly after a whiplash injury. We found that several physical impairments were significantly different between both groups, and that a battery of tests including FRT, PPT, manual palpation over cervical and muscular structures, CCFT, ROM and neck endurance can differentiate between people with acute WAD with or without a headache.

Between group differences were found for the assessment of pain reported during pain sensitivity tests for all measurements including articular and muscular manual palpation and PPT over neural structures except for C3 and C3-C4 assessment. Similar findings have been found for other headache conditions when compared to people without headache [29,42]. However, other tests were not included in our study, such as the assessment of the splenius capitis muscle, which may be of interest in future research for the assessment of these patients. Moreover, tests of pain sensitivity such as pain reproduction are not absolute signs of cervical musculoskeletal dysfunction, since they may just reflect increased sensitization [43]. Widespread pressure pain sensitivity over nerve structures was found in people with tension-type headache, likely reflecting central sensitization [44]. Thus, in people with whiplash-associated headache, the sensitization of the trigemino-cervical nucleus may be underlying the results of our assessments, which may explain why no differences were found for the assessment of C3-C4, which are less related to this nucleus [45]. In that sense, no differences were found when neural tissue mechanosensitivity was assessed. Although it can be present in people with WAD [46], and following or previous hypothesis, the lack of relationship of the tests assessed with the trigemino-cervical nucleus would explain that no differences between groups were found. Although a more complex presentation after a whiplash injury may be related to the presence of neuropathic pain features [47], which may be related to the presence of headache, we did not include people with WAD grade III, as it was done previously [47], so results cannot be compared.

It has been argued that, since pain sensitivity tests may be reflecting a state of increased sensitization rather than a musculoskeletal dysfunction, tests assessing function may be more conclusive [43]. We found that for all movement directions except for extension, significant differences between groups were present. Previously, it was found that reduced cervical ROM was associated with higher neck pain intensity in people with acute WAD [14], and that higher neck pain intensity is more likely in those who present with headache compared to those who don’t [48]. Thus, it may be that since WAD H have higher neck pain intensity, they are more likely to have reduced cervical ROM. Significant differences were also found for the FRT, which is considered an important and reliable test in the assessment and detection of movement impairment of people with headache [49,50]; indeed, the FRT was the only test not directly reflecting pain sensitivity that demonstrated a strong correlation with headache intensity; although causality cannot being established based on our analysis, these findings may suggest the role of upper cervical structures on headache intensity. On another note, no difference was found for posture, measured through the CVA, suggesting that the assessment of posture is not as much related as other tests with the presence of headache.

Neuromuscular dysfunction has been reported to occur in people soon after a whiplash injury [51]. However, no previous studies have assessed whether differences exist between those with and without headache shortly after a whiplash injury. Earlier work has described reduced endurance in people with migraine when compared to headache-free controls which could present with neck pain [52], and impaired CCFT is usually reported as a clinical feature of patients with headache [29,42]. This is the first study however to demonstrate poorer neuromuscular function is poorer when headache is present in acute WAD when compared to the WAD NH group.

A discriminant function analysis found that a group of physical measures composed of neck endurance, manual palpation over cervical and muscular structures, PPT, CCFT, ROM and FRT could discriminate between groups with a sensitivity of 86.7% and specificity of 90%. Discriminant analyses between people with and without headache can provide a comprehensive understanding of the association between different variables and the group membership, providing insight into potential targets to treatment [53], in this case of headache. The diagnostic significance of a single musculoskeletal disturbance has been challenged due to the frequent variability in an isolated measure, such as range of motion [54]. Thus, identifying a pattern of coexisting impairments that demonstrate a high discriminant coefficient may help clinicians to identify which factors determine differences between groups and, therefore, it provides valuable information in relation to which factors may be involved in the presence of headache.

Clinical considerations

We performed a comprehensive assessment of articular, muscular and neural structures, combining self-reported and objective measurements, thus providing a more throughout understanding of physical findings in people with acute WAD and headache. We did not only assess differences in physical examination findings between groups, but also examined the correlation of physical findings with headache intensity and examined which tests could discriminate between groups. In summary, our findings carry clinical implications that extend beyond the identification of the presence or absence of headache soon after a whiplash injury. The identification of specific physical impairments associated with whiplash-associated headache may enhance clinical decision making, and ultimately improve treatment outcome via the identification of impairments which could be targeted to relieve headache for these patients. This study therefore lays the foundation for future research to improve outcomes for patients with whiplash-associated headache.

In the current study we did not attempt to classify the type of headache in those with whiplash-associated headache e.g. cervicogenic headache or tension-type headache. Physical impairments have been identified in different headache conditions such as migraine and tension-type headache in addition to cervicogenic headache, and a recent systematic review [20] found that the only physical impairments which could differentiate between a primary headache and cervicogenic headache, were the FRT and neck strength. Nevertheless, given that cervicogenic headache is very common in people with WAD [55,56], it is very likely that many of the participants presented with cervicogenic headache.

It is also relevant to consider that psychological variables were not included in our analyses, and we therefore may have missed important features especially since the relevance of psychological factors have been demonstrated in people with WAD H [57]. A further consideration was that we collapsed variables for the discriminant function analyses, which could be a source of bias in our results. Moreover, although we assessed manual palpation of the masseter and temporalis muscles, we did not include the assessment of other temporomandibular symptomatology, which may be a limitation since they can be related to whiplash and headache [58]. In addition, the presence of headache could only be a sign of a worse clinical picture. Thus, differences between groups may simply be consequence of increased pain sensitivity, and not a sign of cervical musculoskeletal impairment, as it has been previously discussed; the sensitized trigemino-cervical nucleus in people with headache may increase sensitization of structures innervated by neurons involved in this system, thus leading to an increased sensitization and/or impaired function [43]. Additionally, the sample size may be also a limitation in our study given the large number of variables investigated. We calculated the required sample size based on one of the most frequently investigated physical tests in people with WAD, which was PPT. This sample size allowed us to identify several examination findings which were significantly different between groups. Nevertheless, we acknowledge that future studies should aim to corroborate our findings in larger samples. Finally, the reliability of the physical tests performed should be considered and although previous work has demonstrated moderate to excellent reliability for the majority of tests used [27], some tests demonstrated systematic bias, and this should be considered when interpreting the results.

Conclusions

A wide variety of physical features were significantly different between WAD H and WAD NH soon after a whiplash injury. Some features, namely neck endurance, manual palpation over cervical and muscular structures, PPT, CCFT, ROM and FRT, can discriminate between groups with a sensitivity of around 86% and a specificity of 90%. Future studies should address whether early intervention to modify these physical features can reduce headache in the longer term and which of these findings are more related to an increased sensitization state than a physical impairment.

Biographies

E Anarte-Lazo is a recent PhD graduate from the University of Seville, distinguished for his research on whiplash-associated headache. His work includes numerous publications addressing this issue, contributing to the understanding and treatment of this condition.

D Falla is a renowned doctor in the field of physiotherapy at an international level and the director of the Centre of Precision Rehabilitation for Spinal Pain at the University of Birmingham.

V Devecchi is a recent PhD expert in statistical methods affiliated with the University of Birmingham.

C Bernal-Utrera is a PhD affiliated with the University of Seville with extensive knowledge of the cervical region and numerous related publications.

C Rodriguez-Blanco is a senior PhD affiliated with the University of Seville and the director of the research group CTS954: Innovations in Health and Quality of Life.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Funding Statement

The author(s) reported there is no funding associated with the work featured in this article.

List of abbreviations

WAD

Whiplash-Associated Disorders

HIS

International Headache Society

FRT

Flexion-Rotation Test

CCFT

Cranio-Cervical Flexion

ULNT

Upper Limb Neural Test

NRSC

Numerical Rating Scale over cervical spine

ROM

Range Of Motion

PPT

Pressure-Pain Threshold

FHP

Forward Head Posture

NRSM

Numerical Rating Scale over cervical muscles

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10669817.2024.2372911.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support our findings are available from the corresponding author upon reasonable request.