PROGNOSIS
The best source of subjects for prognostic studies would be all persons in a population who experienced a MVC, which included an acceleration-deceleration event, and thus had the potential to produce a whiplash injury and its clinical manifestations. No reports on prognosis in subjects thus identified were accepted. The next best source for prognostic studies of this problem would be subjects identified by insurance claims with clinical examinations occurring almost immediately after the MVC, or by visits to health-care providers almost immediately after MVC. From sources of this type, information can be gathered very early in the evolution of the disorder on all potential WAD subjects who had been sufficiently concerned to bring themselves almost immediately to either insurance attention or health care. The reports of Norris and Watt, 73 Heise et al, 37 Radanov et al, 82, 85 Burke et al, 11 and Hildingsson and Toolanen 39 fall into this category. Excluded from this type of study are people who did not consider themselves in need of emergency medical attention or of indemnity immediately after the MVC.
Initial presentation
1. Severity: Norris and Watt, 73 in a prospective study of 61 patients presenting to an emergency department after MVC in England, classified subjects based on symptoms and signs at the initial exam. Forty-four percent had symptoms but no reduction in ROM (Group 1), 29% had neck pain and a reduced ROM (Group 2) and 16% had neurologic signs (Group 3). These groupings roughly parallel the Quebec Grades I-III, with the exception that fractures and subluxations could be included in any of the three groups rather than forming a separate group. On X-ray examination, one subject in Group 1 was found to have a fractured transverse process and three subjects in Group 3 had other cervical spine fractures.
Burke et al 11 reported that, of 39 WAD subjects presenting to an emergency department within seven days of MVC, 41% had an initial presentation corresponding to Quebec Grade I, 56% to Quebec Grade II and 3% to Quebec Grade III.
Heise et al 37 also studied patients coming to emergency departments with neck pain after MVC, but did not exclude patients with multiple injuries or bony-tissue injuries of the cervical spine. Of the 155 patients included in their study, 41% had fractures, dislocations, or subluxations of the cervical spine (Quebec Grade IV). The remaining 59% of patients were Quebec Classification Grades I-III.
2. Presenting symptoms: (Table 13). The study of Hildingsson and Toolanen 37 (patients referred to an orthopedic service from an emergency department in Sweden) and secondary analyses of the studies of Radanov et al 82 ("common whiplash" patients in Switzerland identified a mean time of seven days post-MVC) and Norris and Watt 73 (patients presenting to an emergency department in England) revealed that the most common presenting symptoms were neck pain (88% - 100%) and headache (54% - 66%). Other symptoms, such as paresthesias, weakness, dysphagia, visual symptoms, auditory symptoms, and dizziness, varied considerably in frequency, possibly reflecting the different patient sources and health-care system influences.
Among the 92 Grade I-III WAD patients in the study of Heise et al, 37 14 (15%) reported masticatory muscle and temporomandibular joint (TMJ) pain, including one reporting a TMJ click.
3. Radiologic findings: (See Table 14). The radiologic findings at baseline examination were reported directly or ascertainable by secondary analysis in the studies by Norris and Watt 73 and Hildingsson and Toolanen. 39 Half or fewer of the subjects had completely normal cervical spine X rays. Forty-six percent of subjects in the former study and 39% of subjects in the latter study had straight or kyphotic cervical spines. Degenerative spondylosis was noted in 31% of Norris and Watts' and 8% of Hildingsson and Toolanen's subjects. The prevalence of radiologic abnormalities was observed to increase with severity grouping in one of the studies. 73
Prognosis overall (Table 15)
In the study by Norris and Watt 73 the prevalence of every symptom had declined somewhat by six months or more after the MVC. The only exception was visual symptoms, which had been reported initially by only 8% of the subjects. The data, however, were not presented in a way that the total number of subjects reporting any symptoms at follow-up could be reconstructed.
In the early study of Radanov et al, 82 a reanalysis of the data shows that 27% of the subjects were still symptomatic at the six-month follow-up. The distribution of symptoms at baseline was similar to that reported by Norris and Watt, 73 except for paresthesias. The later report of Radanov et al85 described predictors of recovery from headache in subjects with "common whiplash." Trauma-related headache (arising in the occiput and radiating to the frontotemporal region) resolved over time; 35% had headache persisting at three months, and 27% had headache persisting at six months.* This frequency is lower than that reported by Norris and Watt. 73 However, the mean time to follow-up in the Norris and Watt study was approximately two years.
At follow-up, on average two years after the MVC, 39 (42%) subjects in the study of Hildingsson and Toolanen39 reported themselves completely recovered, 13 (14%) reported mild discomfort, and 41 (44%) continued to have major complaints. The authors found no relationship between the potential predictors ascertained at baseline and persisting symptoms at follow-up. By comparison, Rosomoff et al 93 found that 100% of chronic neck pain patients attending a pain clinic had tender trigger points, 54% had a decreased range of motion, 53% had non-dermatomal sensory changes and 15% had rigid contracted muscles. With respect to TMJ symptoms, Heise et al 37 found that at one month, no new symptoms were reported, and that the patients with initial symptoms "reported a decrease in these symptoms." At 1 year, "no new cases of TMJ pain and clicking were reported."
Prognostic factors
In this section, the prognostic importance of single or grouped determinants is described. Overall, the studies of Norris and Watt 73 and Radanov et al 82, 85 both suggest that the severity of the initial injury is a predictor of the persistence of symptoms.
1. Signs and Symptoms: (see Table 16) Radanov et al 82 found that finger paresthesia was predictive of persistence of symptoms for six months after the injury in a series of 78 patients who attended primary care practices within seven days of a whiplash injury. **
In the Radanov et al 85 report examining predictors of persistence of headache in subjects with "common whiplash", the authors conducted a multiple logistic regression analysis to determine predictors of trauma-related headache at baseline, three months, and six months. Initial neck pain intensity, time of onset of neck pain, and depression and well-being scores were associated with headache at baseline. Pre-trauma headache and concurrent severity of neck pain were predictive of headache at three months and at six months. This report suggests that pre-trauma headache and neck pain are associated with a delay in recovery from trauma-related headache after whiplash injury.
The central finding of the study by Norris and Watt 73 is that the presence of musculoskeletal or neurologic signs within three days of a MVC, as indicated by severity group, is predictive of outcome six or more months later. Headache was common across all groups and resolved equivalently in all groups. Neck pain resolved in half of Group 1 subjects, but persisted in the other groups. Paresthesias resolved somewhat by the end of the follow-up in Groups 2 and 3, but not in Group 1. Dysphagia, weakness and dizziness all resolved within six months. Auditory and visual symptoms were rare in Groups 1 and 2, but tended to appear late in the evolution of the disorder. Dizziness was uncommon initially and not present at baseline in Group 1. Table 17 summarizes the core results regarding prognosis within groups classified according to signs and symptoms in this study.
In a small study examining visual complications of whiplash, Burke et al11 followed 39 emergency room subjects for six weeks. At initial exam (less than seven days after the MVC) 10 (26%) had visual symptoms and 16 (41%) had ophthalmologic abnormalities. At the time of follow-up, ophthalmologic abnormalities remained in five subjects of the original 16.
2. Radiological findings: The prognostic importance in WAD of radiologic findings, including osteophytes, end-plate sclerosis, angular deformity and preexisting degenerative changes, was not examined in any accepted studies. *** It should be noted that in the cross-sectional study of Van der Donk et al, 13 X-ray evidence of disc degeneration was associated with nonspecific neck pain in men and not in women, while osteoarthritis was not associated with neck pain.
3. Sociodemographic and economic factors: Sociodemographic and economic factors have been explored in several studies. In the earlier Radanov et al study 82, the groups symptomatic and asymptomatic at six months did not differ in sex, education, injury mechanism, collision fault, or time from injury to initial study examination. Older age was associated with persistent symptoms. All the studies reviewed addressing the influence of financial compensation and legal action on the prognosis of WAD were flawed by substantial selection and information biases. The association of compensation and legal action with outcome in whiplash injury, therefore, remains to be shown.
4. Psychological factors: Several psychological and behavioral factors, such as psychosocial stress, personality characteristics and depressive symptoms have been examined. In the study by Radanov et al, 82 there was no statistically significant association of life history, personality traits and "current psychosocial stress" with persistence of symptoms at six months. Self-report of cognitive impairment, on the other hand, was associated with symptom persistence. However, this study included only 78 patients and therefore had limited statistical power.
Radanov et al 83 report that there was no evidence for differences in attentional processing between "common whiplash" patients (mean duration 24.6 months) and patients with Barré-Léiou syndrome (mean duration 63.5 months), although whiplash patients tended to report more disturbance of cognitive function. The authors suggest that the self-reported cognitive impairment may have been related to the sudden nature of whiplash injury, while in Barré-Léiou syndrome there is a gradual onset, allowing time for adaptation. This suggestion requires further testing.****
Van der Donk et al 113 have suggested that neurotic personality traits are associated with neck pain in the general population. However, this was a cross-sectional study and it is not possible to determine whether the presence of neck pain negatively affected personality or personality was a determinant of persistent neck pain.
In a report addressing a larger number of symptoms in the same cohort, Radanov et al 84 reported that 44% of subjects were symptomatic at three months, 31% at six months, and 24% at one year after entry to the study; 11% were still disabled at three months, 6% at six months, and 4% at one year.
In the Radanov et al 84 cohort, symptoms of radicular irritation were found, by multiple logistic regression, to be predictive of symptom persistence at three, six and 12 month follow-up. Other signs and symptoms predictive of delayed recovery were intensity of initial neck pain, and headache as a result of the current trauma.
Radanov et al.84 report that baseline radiologic evidence of pre-existing osteoarthrosis was predictive of symptom persistence at three months, but not at six or 12 months.
In their later report on a larger cohort, Radanov et al 84 found an association of cognitive disturbances (self-report of forgetfulness and poor concentration), speed of information processing from the personality inventory, and variables interpreted as indicating the intensity of the initial reaction to the trauma (sleep disturbances, score on nervousness), with persistence of symptoms. Contrary to expectation, neuroticism as measured on the personality inventory was associated with earlier recovery.
