Age and Ageing Advance Access originally published online on July 4, 2006
Age and Ageing 2006 35(5):518-520; doi:10.1093/ageing/afl057
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Letter |
Carotid sinus massage – How safe is it?
SIR—Carotid sinus massage (CSM) is a useful diagnostic tool in the investigation of unexplained falls and syncope [1]. Despite the increased use of CSM, concerns persist regarding its safety particularly in older patients. The principal concern relates to the development of neurological events secondary to CSM. The likelihood of performing CSM over atheromatous carotid arteries is higher in older patients with the increased theoretical risk of precipitating an anoxic event or profound haemodynamic changes sufficient to cause a permanent neurological deficit.To date, four major studies [2–5] have reviewed the incidence of neurological complications following CSM. Reported neurological complication rates ranged from 0.17% [4] to 1.0% [5]. Cardiac complications including ventricular [6, 7] and atrial arrhythmias [8] have previously been described only rarely. Data on CSM complications are available from only three centres worldwide, with 79.5% of these patients being from two centres. It is unclear whether CSM complication rates observed at these centres are comparable with those found elsewhere.
Whether differences in patient risk factors account for the apparent 5- to 6-fold variation in complication rates described in these studies is also unclear. This knowledge is important if physicians are to provide meaningful information to their patients and obtain informed consent for the procedure.
Computerised data on CSM studies performed at King’s College Hospital (KCH) (London, UK) and Mid-Western Regional Hospital (MWRH) (Limerick, Ireland) were prospectively collected from August 1995 to March 2004 and January 1998 to March 2004, respectively. A specific field existed in both the databases for operators to record complications arising from CSM as they occurred. The absence of complications was not specifically documented in the KCH database. Data from both databases were collated in a single Excel spreadsheet and further analysed by the author.
Patients were referred from a wide variety of sources including hospital consultants, the Accident and Emergency departments as well as from general practitioners. Indications for CSM were recurrent unexplained falls or syncope together with a clinical suspicion that carotid sinus syndrome may be the underlying diagnosis. Contraindications to CSM were the presence of a carotid bruit, a known carotid stenosis >50%, a history of myocardial infarction or stroke within the preceding 3 months or a history of ventricular arrhythmias or symptomatic bradyarrhythmias [3]. All patients gave informed consent prior to CSM.
Longitudinal massage was applied for 5 s over the point of maximal carotid impulse on the right and then the left carotid artery. A minimum of 30 s was allowed between stimuli. Patients had supine and then erect CSM (using a motorised foot-plate-assisted tilt-table) with continuous beat-to-beat blood pressure (digital artery photo-plethysmography) and heart rate (surface electrocardiogram) monitoring throughout the entire procedure. Devices used included Portapres and Finometer (FMS Medical Instruments, Amsterdam, The Netherlands). Upright CSM was not performed if a clinically significant haemodynamic change was demonstrated while supine.
A neurological complication was defined as a stroke or transient ischaemic attack (TIA). A cardiac complication was defined as any episode of supraventricular tachycardia, ventricular tachycardia or ventricular fibrillation—sustained or unsustained. Age, sex, diagnosis and adverse outcomes were recorded. Patients in whom complications occurred were either admitted for further evaluation or discharged and followed up in medical outpatients as clinically indicated. Using an electronic search of the medical literature, previously published data relating to CSM safety were analysed and compared with our own (Table 1).
|
Comparison of the mean age between KCH and MWRH with the other series was done with one-sample t-tests, taking the series as normative populations.
Comparative data between our study and previously published series are summarised in Table 1. At KCH, neurological complications occurred in two patients at a complication rate of 0.17% (Table 1). One patient had a TIA (Partial Anterior Circulation territory) with expressive dysphasia, right facial nerve weakness and disorientation throughout, which fully resolved within one hour. Another patient sustained a stroke (partial anterior circulation infarct) with confusion, dysphasia and mild right-sided weakness, which resolved apart from residual loss of right hand fine motor function. A CT brain of this patient revealed small vessel disease only. Carotid Doppler scanning revealed heterogeneous plaque with 50% stenoses in both the carotid bulb and internal carotid artery on the left side. The incidence therefore was 0.17% per patient investigated. There were no cardiac complications.
At MWRH, one patient suffered a lacunar stroke with left-sided hemiparesis that resolved apart from a residual left upper limb weakness. An MRI of brain revealed multiple mature infarcts. Another developed nominal aphasia (partial anterior circulation territory) within 24 h of CSM and was persistent beyond 24 h. A subsequent CT of brain revealed atrophic changes only, with no evidence of infarction. One other patient developed dysarthria on tilting upright following supine CSM. This fully resolved within minutes. Carotid Doppler studies were not performed on any of these patients. The incidence of neurological complications was 0.25% per patient investigated. There were two cardiac complications. Both patients developed short non-sustained episodes of ventricular tachycardia. The incidence of cardiac complications was thus 0.17% per patient studied.
The incidence of neurological and cardiac complications per patient studied at both centres combined was 0.21 and 0.08% respectively. The combined rate of persisting neurological deficits was 0.13%.
This study, which to our knowledge is the second largest reported series, reaffirms the safety of CSM. The mean age of our patient population was greater than that of previously published studies (Table 1) for which mean patient age was included. Despite this, there was no increased neurological complication rate.
Comparative demographic and complication rate data between the two study centres and previously published series were broadly similar, with the exception of one series [5] showing an apparently higher complication rate. The differences in complication rates between published studies may be partly explained by heterogeneous recruitment patterns or variable CSM techniques employed by multiple investigators at different centres. This was likely to have been especially prior to the publication of the Newcastle Protocol in 2000 [9]. Different age and co-morbidity profiles as well as retrospective versus prospective analysis of data may also have contributed.
Interestingly, Puggioni et al. [4] did not exclude patients from CSM on the basis of carotid bruits. Neither were carotid Doppler studies routinely performed prior to CSM. Nonetheless, the complication rate was no higher than that shown in previously published data. It also differed from the other studies in that CSM was performed for 10 s rather than the standard 5 s currently recommended by the Newcastle protocol. This did not appear to increase the complication rate (Table 1). The retrospective nature of this study may, however, have underestimated the true complication rate. While every effort was made to ensure the accuracy of our results, we recognise the potential for methodological error in our study, particularly in the underreporting of complications. It is possible that not all complications were prospectively recorded by individual operators as they occurred. Errors may also have occurred when the data from the two centres were collated in a unified database by a single investigator.
Richardson et al. [5] found a higher neurological complication rate than the other studies. The rates of persisting neurological complications were, however, similar in their study and ours, –0.1 and 0.13% respectively. Inconsistent definition between the series of what actually constitutes a neurological complication may partly explain this. We would disagree with the adoption of their definition [5], whereby a neurological complication is defined as ‘any symptom or sign relating to the central or peripheral nervous system’, as this may fail adequately to differentiate pre-syncope and syncope symptoms from stroke and TIA. It is possible that the wide definition given for neurological complications by Richardson et al. [5] led to an artificially high rate compared with other studies.
Our series is the only large series in published literature to our knowledge to describe cardiac complications following CSM. The reasons for this are unclear. Underreporting of cardiac complications in previous studies is a possibility. The co-morbidities and prescribed medications of the two patients who sustained cardiac complications in this study are outlined in Table 2. It is interesting, though perhaps not unsurprising, to note that both patients had pre-existing cardiac disease. While the small complication rate from this study precludes the identification of risk factors, it may be that significant cardiovascular co-morbidity is a risk factor for cardiovascular complications from CSM.
|
This study provides further evidence that CSM is associated with a low risk of either cardiac or neurological complications even in an older population. The data for neurological complication rates are comparable with previously published studies. The apparent differences in reported complication rates between the centres are more likely to be due to differences in study populations and methods than due to true differences in safety.
 |
Key points |
|---|
 Top Key points References |
|---|
- Accurate information on complications following CSM is vital to ensure informed patient consent.
- While there is an apparent difference in complication rates between the published series, it is possible that this is due to methodological variation rather than real differences.
- CSM remains a safe diagnostic procedure even in older population groups.
1 Clinical Age Assessment Unit, Department of Medicine for the Elderly, Mid-Western Regional Hospital, Dooradoyle, Limerick, Ireland
2 Clinical Age Research Unit, Department of Medical Gerontology Bessemer Road, London SE5 9PJ, UK
3 Department of Medical Statistics, King’s College Hospital, Bessemer Road, London SE5 9PJ, UK
* To whom correspondence should be addressed Tel: (+353) 61482623; Fax: (+353) 61485109; Email: thomaswalsh{at}vodafone.ie
|
References |
|---|
|
TopKey pointsReferences |
|---|
- Mcintosh S, da Costa D, Kenny RA. Outcome of an integrated approach to the investigation of dizziness, falls and syncope in elderly patients referred to a syncope clinic. Age Ageing 1993; 22: 53–8.
[Abstract/Free Full Text] - Davies AJ, Kenny RA. Frequency of neurological complications following carotid sinus massage. Am J Cardiol 1998; 81: 1256–7.[CrossRef][ISI][Medline]
- Munro NC, McIntosh S, Lawson J, Morley CA, Sutton R, Kenny RA. Incidence of complications after carotid sinus massage in older patients with syncope. J Am Geriatr Soc 1994; 42: 1248–51.[ISI][Medline]
- Puggioni E, Guiducci V, Brignole M et al. Results and complications of carotid sinus massage performed according to the ‘method of symptoms’. Am J Cardiol 2002; 89: 599–601.[CrossRef][ISI][Medline]
- Richardson DA, Bexton RS, Shaw FE, Kenny RA. Complications of carotid sinus massage – a prospective series of older people. Age Ageing 2000; 29: 413–7.
[Abstract/Free Full Text] - Matthews OA. Ventricular tachycardia induced by carotid sinus stimulation. J Maine Med Assoc 1969; 60: 135–6.[Medline]
- Alexander S, Ping WC. Fatal ventricular fibrillation during carotid sinus stimulation. Am J Cardiol 1966; 18: 289–91.[CrossRef][ISI][Medline]
- Blumenfeld S, Schaeffeler KT, Zullo RJ. An unusual response to carotid sinus pressure. Am Heart J 1950; 40: 319–22.
- Kenny RA, O’Shea D, Parry SW. The Newcastle protocols for head-up tilt table testing in the diagnosis of vasovagal syncope, carotid sinus hypersensitivity, and related disorders. Heart 2000; 83: 564–9.
[Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. B. Keller and L. Lemberg Iatrogenic Sick Sinus Syndrome Am. J. Crit. Care., May 1, 2007; 16(3): 294 - 297. [Full Text] [PDF]
|
|
|
|
 |
|
 Minerva BMJ, October 21, 2006; 333(7573): 866 - 866. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||