Signs and symptoms of vertebrobasilar insufficiency secondary to atherosclerosis: a systematic review

Study identification

A literature search was conducted across six electronic databases: Allied and Alternative Medicine Database (AMED), AgeLine, SPORTDiscus, MEDLINE, Cochrane, and Cumulative Index of Nursing and Allied Health (CINAHL) Plus. The search strategy utilized a combination of Medical Subject Headings (MeSH) terms and free-text terms relevant to VBI and atherosclerosis, including cervical arterial dysfunction, vertebral artery, basilar artery, intracranial circulation, posterior cerebral circulation, cerebrovascular disease, ischemia, atheroma, stenosis, occlusion, clinical features/signs/symptoms/presentation, manifestations, signs, and symptoms. The precise search strings utilized are demonstrated in Table 1. The initial search was undertaken in July 2019 by the first author (CM), and the results were reviewed by the second author (CS); no disagreements occurred between the authors. The search was repeated on four separate occasions between August 2019 and September 2024 to obtain any additional published records over that time.

Study selection

The inclusion and exclusion criteria ensured that all features reported were secondary to atherosclerotic stenosis of the vertebrobasilar territory alone (Table 2). Papers reporting a confirmed diagnosis through the use of the gold standard for the detection of posterior circulation disease and vertebral artery stenosis were included. The invasive technique of intra-arterial digital subtraction angiography (DSA) provides the current gold standard [33], [34], [35], [36]. Noninvasive methods of transcranial Doppler ultrasound (TCD), magnetic resonance imaging (MRI), and computerized tomography angiography (CTA) [35] were also deemed appropriate to inform the diagnosis. To allow a full scoping review of the current literature, the search was open to all study designs with no exclusions.

Given the relevance to manual-treatment screening and examination, it was vital that the reported symptoms presented prior to permanent neurological damage and occurred in the absence of other pathological mechanisms. Cases reporting collateral circulatory pathology, presence of neurological infarcts (previous or existing) at the time of the reports, or the presence of other pathological causes – ie, thromboembolic source – were excluded. Cases under medical management were included, because they were deemed pathologically consistent and practically relevant.

Screening process

Results from the database search were manually and systematically screened against the inclusion/exclusion criterion; following the exclusion of duplicates, an independent reviewer also verified the study selection. Reference lists of included articles were manually screened against the same criteria; the final selection was analyzed, and relevant data were extracted.

Data extraction and analysis

To ensure consistent and reliable data extraction, a bespoke form was devised. The following information was systematically extracted and tabulated: authors, year of publication, study design, length of study, the number of cases/patients, diagnostic technique, comorbidities and risk factors, and symptoms and their characteristics (duration/frequency/stenosis rates). A second independent reviewer analyzed a sample of articles against the bespoke extraction form to ensure reliable data gathering. It was agreed by both reviewers that the extraction process was suitable and permitted the capture of all relevant information. A meta-analysis was not performed due to the substantial clinical and methodological heterogeneity of the included studies. A narrative synthesis was performed with results summarized descriptively.

Assessment of methodological quality

The methodological quality of the included studies was evaluated systematically utilizing the appropriate critical appraisal tools [37] based on the article study design, which in this instance were those designed for case controls, case series, and case reports. The Joanna Briggs Institute appraisal tools were utilized to provide a detailed blueprint to follow during the critique, thus allowing the rigorous assessment of the included articles to determine the levels of possible bias in each study and the extent to which these factors were addressed.

Study selection

See Figure 1 for the selection and screening process. Following Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines [38], the database search identified 283 articles. After the removal of duplicate articles (n=33) and excluding articles failing to meet the research criteria plus articles agreed to be unsuitable by independent reviewer (n=1), 15 studies remained, in which 93 individual cases were reported and information was systematically extracted (Table 3). The patient demographics were 50 male and 43 female cases, and the average age was 64 years old (SD=9), ranging from 39 to 81 years old.

Figure 1:

A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram, showing the systematic research methodology and screening process.

Study characteristics

Of the 15 articles, three were published within the last decade and six were published within the last 20 years. The study designs utilized for the manuscripts were case series (n=9), case reports (n=5) and case control studies (n=1). Cases either explored revascularization techniques in those with severe atherosclerotic stenosis who had failed medical intervention or who investigated patients presenting with features of vertigo/dizziness. The cases provide clear symptom reports; however, 10 failed to present sufficient detail regarding symptom duration, frequency, and progression. Case series report information regarding stenosis rates (%) present in the vertebral and basilar arteries provided additional anatomical insight into the atherosclerotic process.

To confirm the clinical diagnosis of VBI, nine different imaging techniques were utilized, and in many cases, multiple methods were utilized. In 55 cases, it was not possible to identify the precise frequency of methods utilized to confirm the diagnosis; thus, these data were not included and have no bearing on the following information. The three commonly utilized diagnostic imaging techniques were MRI (29.8 %), the gold-standard DSA (21.1 %), and CTA (12.3 %); see Table 4. In total, 37 different symptoms were reported; either presenting in isolation or combination. Original symptom reports can be found in Table 3. Vertigo (27.7 %), unsteadiness/incoordination (7.4 %), and diplopia (7.0 %) were the most frequently reported symptoms of VBI. Symptoms associated with 1) vertigo and dizziness (including posture induced vertigo, syncope), 2) visual dysfunction (diplopia, illusions, hallucinations, field defects, complete vision loss and blurred vision) and 3) cognitive dysfunction (confusion, loss of consciousness, memory disturbance, mental status change, dysarthria, dysphagia and drop attacks) accounted for 32.5 % (n=88), 17.7 % (n=48) and 15.5 % (n=42) of reports, respectively (Figure 2).

Figure 2:

Frequency of symptom of vertebrobasilar insufficiency (VBI) grouped by category, as a percent of the total reported (n=271).

The general characteristics of reported symptoms, i.e., duration, frequency and progress over time, range extensively throughout the results with no clear pattern apparent; the duration of symptom episodes ranged from 1 min to 40 min [6], 41], 45], to 2–10 h [49]. The frequency of episodes ranged from one to seven times daily [41]; among cases reporting a 3-month history of VBI episodes [40] to 2–3 year history of preceding symptoms [41], 44]. One study reported 200 fluctuating episodes over a 2-year period [40], while another indicates resolution of symptoms after 14 days from the initial onset [41]. Common risk factors reported by cases (n=67) included hypertension 40.3 %, diabetes mellitus 19.4 %, and smoking 13.4 % (Table 5).

Methodological quality

The included studies demonstrated moderate to good levels of reliability. Levels of research evidence ranged through observational analytic designs and descriptive studies (Levels III-IV) in accordance to The Joanna Briggs Institute Critical Appraisal Tool [37]; limitations inevitably exist across case series/reports. Due to small sample sizes, 7 out of 15 reporting a singular suitable case results (Table 3; Article no. 1, 3, 8, 10, 12–14) lack research power and provide little basis to generalize findings to the wider population. Selection bias also limits the wider application of results because the sample selection limited by the characteristics of each study unavoidably limits the external validity.

All cases utilized standardized and validated methods to confirm VBI; patient demographics and clinical information were sufficiently presented. The inclusion criteria were not explicit across all cases series, as complete inclusion (Table 3; Article no. 6, 7, 9, 13, and 15) and consecutive inclusion (Table 3; Article no. 6, 11, and 13) were reported. In the remaining cases, the inclusion criteria were unclear, reducing the reliability of results.

Those studies observing patients over a follow-up period of time to monitor symptoms or stenosis rates generally demonstrated a large variance in quality; either the protocol was not standardized and inconsistently measured (n=5), the data were not collected for all patients (n=1), and/or the short-term effects were recorded (n=1), potentially damaging the internal validity and reliability of outcomes. Given the primary objectives and outcome measures across cases – ie, stenosis rate, hemodynamic flow, symptom resolution – statistical analysis was not performed in 14 cases. The case control study [12] utilized independent blinded assessment of radiological images to improve internal reliability; outcome measures were standardized across comparison groups. The control group was only utilized as a baseline comparison for vestibular testing, and the study was unable to measure perfusion rate.

Clinical findings

The aim of this systematic review was to summarize the reported features of atherosclerotic-induced VBI. In turn, this will provide evidence to help clinicians successfully identify atherosclerotic-induced VBI, aid differential diagnosis, and guide the use of manual therapy interventions.

In the literature reviewed, 37 different symptoms were reported. Vertigo was the most common. The symptoms inclusive to Coman’s cardinal signs accounted for 22.2 % of all symptoms (Table 6). A large percentage of symptoms reported in the literature are not sufficiently represented through the use of Coman’s cardinal signs, perhaps indicating a risk of misidentification of VBI and therefore potential clinical mismanagement. Findings suggest that the current literature does not support the sole use of Coman’s cardinal signs in the clinical detection of VBI secondary to atherosclerosis. Further to this, it perhaps suggests a need to re-evaluate the terms constituting Coman’s 5D’s because these are still inherent within medical and manual treatment (physiotherapy, osteopathy, chiropractic) education and teaching. Results here imply that relying on Coman’s 5D’s in isolation is not sufficient to reliably raise clinical suspicion and inform the safe use of manual therapy intervention. Musculoskeletal and manual therapists need to place large value upon the ability to take an accurate patient history in order to build a reliable clinical picture in which to draw clinical conclusions.

The commonality of vertigo as a symptom of VBI raises questions as to whether evidence is sufficient to advance the information provided within the Coman’s cardinal signs and whether “vertigo” warrants a greater degree of acknowledgement. Having said that, there needs to be an element of caution. Within the literature, there appears to be an inconsistent use of terminology and an unclear distinction between terms associated with vertigo, and operational definitions are rarely provided. Due to the ambiguity of whether reports of dizziness constituted vertigo or lightheadedness, this review kept the terms separate. A lack of precision in research terminology was noted by Kerry and Taylor [52], who further emphasized the importance of differentiating the nature of dizziness in order to establish whether a vascular or nonvascular cause is present. Misinterpretation across the literature may lead to incorrect representation and carryover of inaccurate information, potentially compromising the validity of this review.

The high incidence of vertigo may reflect the selection bias present within the studies; combined with the retrospective nature and study heterogeneity present, this suggests that caution should be made when drawing conclusions. The case series described by Grad and Baloh [6] accounted for 62 % of vertigo reports, thus possessing far more weighting on results comparatively to individual cases. Excluding all cases from Grad and Baloh [6] makes little change in the overall results; the comparative margins between symptom frequencies are altered, yet vertigo remains the most common symptom.

Vertigo as a feature of VBI

Four cases reported isolated episodes of vertigo and dizziness secondary to atherosclerotic stenosis in the posterior cerebral circulation (Table 3; Article no. 2–5). Previous understanding [10], 11], 52] suggested that vertigo should not be attributed to VBI unless it is in the presence of other neurological signs in the brainstem. However, results of this review in addition to previous reports [13], 53] provide contradicting evidence that episodes of isolated vertigo can be featured in VBI induced by posterior circulation ischemia.

Risk factors of VBI

Diabetes and smoking were identified as common risk factors, with hypertension (40.3 %) being the most frequent [54]. Coronary artery disease and hyperlipidemia were also noted, which also increases the risk of VBI [54]. The results of this study resembles reports in the ‘New England Posterior Circulation Registry’ outlined by Ishiyama and Ishiyama [9]. However, a volume of data remains unaccounted for, as the information was either not reported or could not be extracted accurately in 49 % of the cases. Nevertheless, a strong link between risk factors and vertebrobasilar ischemia exists and assists in identifying those at risk of VBI through vascular compromise.

Research restrictions and limitations

Volumes of literature exist containing information that could better inform results and improve study reliability. Regrettably, in 20 of the studies reviewed, transient VBI reports were grouped with neurological infarct data that could not be extracted and were therefore excluded from this paper. Cases investigating arterial revascularization techniques lacked detail describing symptom pattern and timelines preceding the intervention. Given the objectives of such studies, the lack of information was understandable; however, a potential opportunity to build the clinical picture of VBI may have been missed. Symptom duration is particularly useful regarding dizziness and vertigo, because duration can help determine whether symptoms are caused by transient ischemia or a benign inner ear disorder [41].

Cases in this review were subject to selection bias due to their retrospective design, ultimately affecting the review in many ways. Although information generally was systematically retrieved, missing or incomplete information was a common issue. Results highlight the sporadic and inconsistent pattern of VBI features and potential progression, and the small number of accurate reports limits the reliability and external validity of this information. Studies included patients who typically failed medical management and possessed high vertebrobasilar stenosis rates. Thus, there is a gap in the literature regarding those with mild or moderate arterial stenosis who are not under medical management. Studies reported presenting episodes of VBI across a time span of 3 months to 3 years, indicating that features of VBI may occur in earlier stages of vertebrobasilar atherosclerosis. Information on this patient population in which VBI has not yet been identified is very relevant. Selection bias exists toward patients specifically presenting with vertigo; thus, the conclusions drawn may not be reliably applicable to those other cases in which vertigo is not a prominent feature.

Recommendations for future research and practice

Resolution toward distinguishing the features of vertigo and other forms of dizziness is warranted, and future research should provide clear definitions alongside the reports. The need for clarity is outlined and exemplification is provided in this review and in previous studies [9], 52], 53]. Considering that vertigo as a singular phenomenon separate from dizziness is perhaps justified; however, given the common practical use of Coman’s cardinal signs, it may be more pragmatic to distinguish the terms through clinical education and emphasize the distinct characteristics of each. Additionally, improving the knowledge of known arterial presentations and the diverse features of VBI could strengthen the clinician’s differential diagnostic skills, a notion also proposed by Kerry et al. [28].

To help develop a stronger evidence base, future clinical case studies reporting VBI should present greater levels of relevant clinical data and clearly differentiate the symptoms relative to transient ischemia, neurological infarcts, and arterial dissection. Reports should collect accurate information pertaining to the pattern, duration, and timeline of VBI symptoms. Such information would greatly benefit differential diagnostic processes. The future use of cohort study designs (prospective observational studies) tailored toward the retrieval of symptom pattern data may permit complete data collection; the need for larger sample sizes and the risk of attrition bias would need to be considered to avoid compromising the validity of the potential results.