Causes of Persistent Dizziness in Elderly Patients in Primary Care

Maarsingh et al’s report on the causes of persistent dizziness in the elderly in primary care is very impressive. The authors evaluated the subjects prospectively, using an extensive, uniform set of diagnostic tests. The wide variety of causes of dizziness in these patients, and the large number of contributing causes per patient, can be overwhelming for the primary care provider. In the US, primary care physicians typically have, at most, twenty- minutes per patient visit. It would be helpful if the authors would next study and report on a practical approach to the office evaluation of the dizzy elderly.

The authors’ comment that the semantic and cultural meanings of the word “dizziness” may differ in Dutch and English was intriguing. As an English speaker, I would be interested to learn more about the cultural meanings of dizziness for the Dutch.

Competing interests:   None declared

This is an interesting article describing a process in assessing dizziness in elderly by three family physicians assisted by an expert system. As I mentioned in one of my publications, the patients referred to a specialized clinic, such as ours, belong to a selected population, suffering from dizziness, although I suspect that these are the most difficult cases in reaching a diagnosis by a family physician. No question heart disease is prevalent in aging and it is tempting to attribute all sorts of symptoms and complaints to it. However frequently in dizziness, which is a symptom and not a disease, the question arises whether heart disease is the cause or there are other underlying less obvious but actual causes.

In my experience, after evaluating more than 24,000 patients complaining of dizziness during more than 60,000 visits, there are processes to help in solving such a difficulty. First, to include in the physicians group, specialists, such neurologists and otolaryngologists, interested in this particular problem, and second to follow up the patients, especially in view of the fact that frequently dizziness is characterized by remissions and recurrences. It is of interest to mention that the authors, to my satisfaction, discuss these points! In a number of cases I changed my mind when I reassessed a patient during a time period the dizziness was active.

In closing, I would like to emphasize that there is still a lot more to learn about this frequent complaint in all ages, keeping in mind that a specific syndrome called "dizziness in aging" does not exist, at least in my experience. All in all, this is an excellent piece of work and the authors have to be congratulated.

A. Katsarkas, M.D., M.Sc., FRCS(C). Professor, Dept. of Otolaryngology, McGill University. Director, Dizziness Clinic, Royal Victoria Hospital.

Competing interests:   None declared

I would like to thank Termeer and colleagues for critically reading our article[1] and for their time and effort to participate in the online discussion.[2] I have a few remarks, though.

First of all, it is tempting to consider a case-control study to be the solution for the diagnostic dilemmas dizziness confronts us with. However, it is not. Even if a diagnostic test reveals a significant difference between dizzy patients and controls, for example limited neck movement in 52% of dizzy subjects vs. 36% in controls,[3] we do not know if the target condition (in this example neck disease) actually is a contributory cause of dizziness, because we lack information about the diagnostic accuracy of this test. At this moment, this is the state of the science: a) many diagnostic tests for evaluating dizziness have been described in observational studies, but only few tests have been evaluated in a diagnostic accuracy study (mainly tests for neuro-otological conditions), b) of this limited amount of diagnostic accuracy studies, almost all studies were conducted in a secondary/tertiary care setting, and c) almost no diagnostic accuracy study on dizziness included a spectrum of patients representative of primary care patients.[4] As a result of this huge lack of evidence, the interpretation of results of diagnostic tests performed among unselected dizzy primary care patients is delicate business. Panel diagnosis is a widely accepted methodological instrument to deal with this diagnostic uncertainty[5] and in my opinion (and others)[6] superior to most previous studies that have relied on a single clinician to determine cause. Although a case-control study does not solve diagnostic dilemmas, it definitely may widen the perspective on dizziness. Therefore, our research group DIEP (Dizziness In Elderly Patients) actually did perform the same diagnostic evaluation in a control group [Dros/Maarsingh et al., work in progress], next to other study designs, like principal component analysis [Dros/Maarsingh, submitted], a follow-up study [Dros/Maarsingh, work in progress], and a prediction study [Maarsingh/Dros, submitted].

Second, the reference of Termeer and colleagues to the STROBE checklist is a bit misleading, as it suggests empirical support for the observations being made. However, the STROBE Statement is a checklist for reporting (i.e. which items should be included in reports of observational studies?), a topic Termeer and colleagues hardly discuss. Moreover, the reporting of the results of our study actually meets all the items of the checklist.[7]

Third, I agree that our panel lacked diagnostic criteria for items of the medical history. However, this was an explicit choice, as such criteria are hardly described (only for specific neuro-otologic conditions) and they lack empirical evidence.[4]

Fourth, Termeer and colleagues were surprised by the high percentage of patients willing to participate, referring to the study of Van der Wouden et al. (Lasagna’s law).[8] However, the primary outcome of the study of Van der Wouden et al. was not “willingness of patients to participate”, but “success of recruitment, measured by the number of patients recruited as proportion of the number that was planned”. Nevertheless, the high percentage of participating patients may be explained by our effort to search the electronic databases of all practices each month in order to identify dizzy patients family physicians had failed to invite (in total 15% of the initial study population). Additionally (although not measured), I had the impression our study population was highly motivated to participate.

Finally, Termeer and colleagues were worried about underestimation of visual and hearing impairment as a contributory cause of dizziness. However, we initially excluded one patient with severe visual impairment (definition: corrected visual acuity of <3/60 in the best eye), and no patients with severe hearing impairment (definition: verbal communication impossible), so the underestimation of these causes of dizziness is probably negligible.

Reference List
(1) Maarsingh OR, Dros J, Schellevis FG et al. Causes of persistent dizziness in elderly patients in primary care. Ann Fam Med 2010; 8(3):196-205.
(2) Termeer EH, Van Dijk WD, Eveleigh RM. Dizziness, a brain teasing symptom http://www.annfammed.org/cgi/eletters/8/3/196. Ann Fam Med 2010; 8.
(3) Colledge NR, Barr-Hamilton RM, Lewis SJ et al. Evaluation of investigations to diagnose the cause of dizziness in elderly people: a community based controlled study. BMJ 1996; 313(7060):788-792.
(4) Dros J, Maarsingh OR, Van der Horst HE et al. Dizziness in primary care: a systematic review of diagnostic tests. CMAJ 2010. In press.
(5) Rutjes AW, Reitsma JB, Coomarasamy A et al. Evaluation of diagnostic tests when there is no gold standard. A review of methods. Health Technol Assess 2007; 11(50):iii, ix-51.
(6) Kroenke K, Lucas CA, Rosenberg ML et al. Causes of persistent dizziness. A prospective study of 100 patients in ambulatory care. Ann Intern Med 1992; 117(11):898-904.
(7) http://www.strobe-statement.org/index.php?id=available-checklists. 2010.
(8) van der Wouden JC, Blankenstein AH, Huibers MJ et al. Survey among 78 studies showed that Lasagna's law holds in Dutch primary care research. J Clin Epidemiol 2007; 60(8):819-824.

Competing interests:   None declared

According to the final step in the RADICAL approach as proposed by this Journal, we would like to share our main points of discussion on the article by Maarsingh and collegues, in part based on the rules of the STROBE Statement (1).

We would like to start with complimenting the authors with this interesting article as it emphasizes again the need for a broad and multifactorial approach towards this complex problem commonly observed in general practice (2). Although, we do think that the conclusion stating that cardiovascular disease was found to be the most common major cause of dizziness in elderly, is a bold statement. The authors put much effort into constructing a valid design, that represents the general population. Unfortunately, the result appears rather complex itself, their observational approach lacks a comparable control group and innate to the design of this study, several assumptions had to be made and disagreements had to be overcome. Moreover, the claimed causes and subtypes of dizziness were democratically based on subjective judgments of three experts instead of pathophysiologic findings, resulting in low interrater agreements. (It is noteworthy that despite the low interrater agreement, consensus for dizziness subtypes was achieved in 96%). For these reasons, we suggest the authors to be more reserved in their conclusions.

In addition to these limitations, we are eager for more interpretations regarding the study results. They achieved to clearly describe both dizziness subtypes and contributing causes, so we hoped to get a glance of how these two correlate and how this could guide us in diagnosis - 21 options as from their Delphi study – as well as treatment.

Apart from these main issues, we wish to address two other remarks. First, the assignment of possible causes were said to be based on medical history, physical examination and additional tests. The diagnostic criteria as from appendix 2 help to guide in contributory causes. Although good efforts were made to restrict investigators bias, they unfortunately used medical history as their key diagnostic tool, which is not embedded in these criteria. In addition, they underestimate (severe) visual and hearing impairment as a contributing cause, since these were excluded from study entry.

Second, the high percentage of patients willing to participate amazed us: 4% from all eligible patients asked to participate were unwilling or failed their appointments. Other studies executed in family practice show percentages of participants around 70%; this pitfall in ineffective recruitment is indicated by Lasagna’s law (3). For this reason, we wonder if we should doubt their initial number of approached patients, and hence their generalizability, or if their total diagnostic work-up substantially improves participation.

In conclusion, the multifactorial character of dizziness makes it an interesting, but complex symptom for the physician and the researcher alike.

References:
1) http://www.strobe-statement.org/index.php?id=available-checklists
2) Maarsingh OR, Dros J, Schellevis FG, van Weert HC, van der Windt DA, ter RG, et al. Causes of persistent dizziness in elderly patients in primary care. Ann Fam Med 2010 May;8(3):196-205.
3) Van der Wouden JC, Blankenstein AH, Huibers MJ, van der Windt DA, Stalman WA, Verhagen AP. Survey among 78 studies showed that Lasagna's law holds in Dutch primary care research. J Clin Epidemiol 2007 Aug;60(8):819-24.

Competing interests:   None declared