Research Review By Dr. Michael Haneline©

Date Posted:

January 2010

Study Title:

Validity of the Doppler velocimeter in examination of vertebral artery blood flow and its use in pre-manipulative screening of the neck


Thomas L et al.

Author's Affiliations:

Discipline of Physiotherapy, University of Newcastle, Callaghan, NSW, Australia.

Publication Information:

Manual Therapy 2009; 14: 544–549.

Background Information:

Although the precise rate is not known, stroke appears to very rarely be associated with cervical spine manipulation (CSM). Researchers have hypothesized that the inner lining of the cervical arteries, particularly the vertebral artery (VA), may become injured during CSM leading to dissection and consequent formation of thrombi, which may dislodge and become emboli.

Current thinking suggests that there must be an underlying condition that weakens the artery and makes it susceptible injury. Otherwise there would be innumerably more cervical artery dissections, since most people are exposed to potentially injurious neck movements every day, but only a very small minority of them ever develops signs or symptoms.

A variety of clinical screening tests have been designed to identify those patients who are at risk of developing vertebrobasilar insufficiency (VBI) following CSM. However, the validity of these tests has been questioned because they lack both sensitivity (the ability of a test to correctly identify those who have the condition) and specificity (the ability of a test to correctly identify those without the condition).

One such screening test involves the use of a hand-held portable continuous wave Doppler device (velocimeter), which can be used to examine the patency of the VAs. This device has been reported to have fairly high inter-rater reliability (0.78) and 100% sensitivity and specificity for identifying “persistent” or “major reduction” in VA blood flow. In previous studies, the results of the velocimeter tests were compared with the findings from duplex ultrasound, which is considered to be a criterion standard. The “gold standard” for detecting patency of the cervical arteries is magnetic resonance angiography (MRA), but it is not practical as a screening test.

The current study had two objectives:
  1. to determine the interexaminer reliability of trained physiotherapists in detecting VA blood flow abnormalities using the Doppler velocimeter, and
  2. to determine the sensitivity and specificity of the procedure by comparing its results to duplex ultrasound tests.

Pertinent Results:

A total of 60 participants were recruited into the study, including 19 males and 41 females. Ages ranged from 21 to 71 years (mean 45.5 years). Only 57 of the participants completed the velocimeter examination because the examiners were unable to obtain satisfactory Doppler readings on 2 of them and 1 re-located and was consequently lost to follow up.

Pertinent Findings:
  • The specificity of the velocimeter examination ranged from 0.78 to 0.88, which is considered to be fair to good.
  • The test’s sensitivity ranged from 0.25 to 0.38, which is considered to be poor.
  • Intraexaminer reliability was also considered to be poor, with reported kappa scores that ranged from 0.15 to 0.26. A kappa score of 0.40 is the minimum value that can be considered acceptable for most clinical tests.
The participants in this study were “healthy volunteers” who were excluded if they had certain cervical spine abnormalities. Studies that assess the validity of diagnostic tests should include a spectrum of patients, some who do not have the condition under study and some who have the condition in varying degrees of severity. The “condition” was abnormal blood flow in this case, which was present in about 30% of the participants.

Clinical Application & Conclusions:

The results of this study do not support the use of the velocimeter in detecting abnormal VA blood flow in the upper cervical spine in the neutral or full head rotation positions.

When considering the validity of a test (i.e. whether it is in fact capable of measuring the characteristic it was designed to measure), one must keep in mind that a test that has unacceptably low reliability can never be considered valid. Therefore, based on the results of the interexaminer reliability portion of this study, the premanipulative screening via velocimiter cannot be considered valid, even though its sensitivity was in the acceptable range.

The reasoning behind labeling people with only one fully patent VA as being at higher risk for stroke following CSM is that if the other artery becomes occluded, there would be a reduced ability for collateral circulation to supply blood to the brain. This idea was first proposed by Haynes, who based it on previous studies in which major restriction in the VA blood flow during contralateral cervical rotation occurred significantly more often in patients who had a history of vertebrobasilar transient ischemic attacks. (1)

A caveat that was mentioned by the authors is that it is unknown whether alterations in blood flow are correlated with a greater risk of developing VA-related stroke following CSM. They went on to point out that VA blood flow was abnormal in some of the study participants, yet none of them had symptoms or signs of VBI. The authors surmised that such VA blood flow abnormalities may actually be normal variants and that the normal blood flow to the brain is maintained by collateral circulation.

One is left somewhat confused by this discussion, however, because the authors went on to relate the logic of Haynes in that “…marked reduction or loss of VA blood flow on neck rotation… should be considered a contraindication to neck manipulation as a manipulative thrust might cause further stress of the artery resulting in mechanical damage to the vessel.”

The validity of the velocimeter for the described purpose was not supported in this study, but it prompted 2 letters to the editor that pointed out what they considered to be serious limitations, casting doubts on the authors’ conclusions. (See “Study Strengths/Weaknesses” section below)

Study Methods:

Participants were recruited through advertisements; however, the means of recruiting the 3 physiotherapists who participated were not described.

The physiotherapists each had at least 5 years of clinical experience in manual therapy and were familiar with the Australian Physiotherapy Association Clinical Guidelines for Pre-manipulative Procedures for the Cervical Spine. They were trained in the use of the velocimeter in a 3 hour workshop where they were briefed on the experimental procedures and were able to practice the examination technique while under supervision. The examiners were then given velocimeters and asked to use them 3 to 4 times per week over the next 2 months. Ultimately, they were assessed to make sure they met minimum standards of competency.

Exclusion criteria:
  • cervical spine instability
  • inflammatory disease
  • uncontrolled cardiovascular disease which would normally contraindicate cervical manipulation
Participants were first examined via duplex ultrasound by a qualified ultrasonographer to measure VA blood flow, followed by examination via a Doppler velocimeter by 3 trained physiotherapists. Resting heart rate, blood pressure, and cervical range of motion were also measured.

The examiners performed the velocimeter evaluations in random order, but it is not clear whether they conducted the examinations independently and without feedback from the subjects. Both VAs were examined at the level of the atlantoaxial segment with the probe directed to the sub-occipital level posterior to the mastoid process.

Study Strengths / Weaknesses:

In studies that assess the reliability and/or validity of diagnostic tests, the examiners should be blinded as to each others’ results. This is because when an examiner knows what the others found, especially the results of the comparison test (Duplex ultrasound in this case), it has the potential to greatly influence their findings. It was not clear whether blinding was utilized in this study though.

The subjects should have been blinded from the examiners’ findings, but the topic was not addressed.

The information that was obtained from the George et al. 1981 reference is outdated and should not be used to decide whether a patient is a candidate for CSM.

An important component of a research article is a section that addresses study limitations. This section should always be included because all studies have weaknesses. Nevertheless, one was not included here.

Haynes, who did some of the early work that this study was based on, criticized the methods that were used in a letter to the editor for several reasons:
  1. The authors used suboccipital duplex scanning as a comparison, but there is no evidence in the literature to support its validity. This was also a criticism listed in another letter by Karl (4).
  2. The percentage of subjects who had a positive duplex examination (23.28%) was disproportionally higher than previous reports (less than 4%).
  3. Haynes trained the author of this study on the velocimeter technique using only 3 example patients, which he considered to be inadequate. This person went on to train the participating physiotherapists.
Haynes mentioned that “The authors’ conclusion ‘… use of the velocimeter as a premanipulative screening tool cannot be recommended at this time’ lacks legitimacy due to the serious limitations of their study with aberrant results.”

Additional References:

  1. Haynes M. Vertebral arteries and cervical movement: Doppler ultrasound velocimetry for screening before manipulation. Journal of Manipulative and Physiological Therapeutics 2002; 25(9): 556–67.
  2. George PE, Silverstein HT, Wallace H, Marshall M. Identification of the high risk pre-stroke patient. Journal of Chiropractic 1981; Suppl: 26–8.
  3. Haynes, M. Reply to Thomas LC, Rivett DA, Bolton PS. Validity of the Doppler velocimeter in examination of vertebral artery blood flow and its use in pre-manipulative screening of the neck. Man Ther 2009; 14(5): e6; author reply e7-8.
  4. Karl, V., Comments on the article by Lucy C. Thomas et al. "Validity of the Doppler velocimeter in examination of vertebral artery blood flow and its use in pre-manipulative screening of the neck", Manual Therapy 2009; 14(5):544-9. Man Ther, 2009. 14(6): p. e17; author reply e18.