| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* From the Department of Medicine, Cooper University Hospital, University of Medicine and Dentistry of New Jersey/Robert Wood Johnson Medical School, Camden, NJ.
Correspondence to: Toby R. Engel, MD, 1 Cooper Plaza, Department of Cardiology, Camden NJ 08103; e-mail: engel-toby{at}cooperhealth.edu
Abstract
Background: Syncope or sudden death has been associated with a short QT interval (QTc, < 300 ms), the so-called short QT syndrome. The current prevalence of this syndrome is unknown. The aim of this study was to evaluate the prevalence of short QT intervals (ie, QTc, < 300ms) in a general hospital population.
Methods: We retrospectively queried 479,120 consecutive ECGs that had been archived (Marquette MAC 5000 Resting ECG System; GE Healthcare; Boston, MA) over a 16-year period. We examined the distribution of QT intervals in our population from 150 randomly selected ECGs with normal findings, excluding patients who had been receiving medications known to prolong the QT interval.
Results: From 1988 to 2004, 479,120 ECGs from 106,432 patients were analyzed, which reported 215 tracings with a QTc of < 300 ms. Each ECG was then measured manually, and no QTc of < 300 ms was validated (67% were found to be in error because of a pacemaker artifact, 17% showed supraventricular tachycardia with inaccurate detection of the T-wave offset, and 16% were found to have an error in the cycle length calculation). Therefore, not one of the 106,432 patients was found to have a QTc of < 300 ms. The mean QTc (± SD) was 430 ± 19 ms (95% confidence interval, 392 to 468 ms). The QTc of < 300 ms would then reflect > 5 SDs shorter than the mean QTc.
Conclusions: The short QTc reported by an ECG computer was inaccurate and required manual correction. Short QT syndrome, defined as a QTc of 
300 ms, is rare. We were unable to find one patient among a population > 100,000 patients with a true QTc of < 300 ms.
Key Words: arrhythmias • ECG • electrophysiology • QT intervals • sudden death
In less than 10% of cases, sudden death occurs in patients with structurally normal hearts, mainly due to alterations in repolarization (eg, long QT syndrome).1 Sudden death has also been described in patients with very short QT intervals, the so called short QT syndrome.2 Several families with short QT syndrome have subsequently been reported.3456 It is possible that short QT syndrome as an explanation for unexplained sudden death has often been overlooked. In this regard, we examined the prevalence of short QT intervals in a hospitalized population in order to detect a possible relationship between extremely short QT intervals and arrhythmic events.
Materials and Methods
We retrospectively queried 479,120 consecutive ECGs that had been archived (Marquette MAC 5000 Resting ECG System; GE Healthcare; Chalfont, St. Giles, UK) at our hospital between 1988 and 2004. The ECG analysis program requires data that have been acquired at a rate of 500 samples per second. The standard filter rate was set at 150 Hz for the low-pass filter without a high-pass filter. The first step in the ECG analysis program (Marquette 12SL ECG analysis program; GE Healthcare) is the identification of each QRS complex. After the filtering is complete, the program attenuates both low-frequency and high-frequency waves, leaving mid-band frequencies for interval detection.
The resting ECG system (MAC 5000) measures the interval in each of the 12 leads from the earliest detection of depolarization based on a high-frequency signal in any lead to the latest detection of repolarization in any lead. The onset and offset of the T wave is measured by an analysis of the simultaneous slopes in all 12 leads. The QT interval is then corrected for heart rate using the Bazett formula (QTc = QT interval/
R-R interval).7 ECGs with a QTc of < 300 ms were then manually checked by the authors at 25 mm/s paper speed, using lead II, as most normal reference ranges are based on this lead.8
This study had the following three components: (1) to assess the prevalence of the QTc of < 300 ms in 479,120 ECGs; (2) to evaluate the accuracy of automated QTc measurements compared to physician measurement when the QT is short; and (3) to assess the distribution of the QTc in a sample of patients from our general hospital population.
300 ms and do not use an uncorrected QT in the definition of the syndrome.9
Statistical Analysis
The results were expressed as the mean ± SD. A paired Student t test (two-sided) was used to compare values. A p value of < 0.05 was considered to be statistically significant.
Results
The 479,120 ECGs examined were acquired from 106,432 patients. The resting ECG system (MAC 5000; GE Healthcare) reported 215 tracings from 138 patients with a QTc of < 300 ms. These QT intervals were then manually controlled by the authors, and no QTc of < 300 ms was validated (ie, all were in error). The majority of computer errors (67%) were due to a pacemaker artifact with the computer mistakenly using pacemaker stimuli for the calculation of the QT interval. Another cause of error (17%) seemed to be supraventricular tachycardia with incorrect selection of the T-wave offset, usually because of flutter waves. The remaining 16% of errors were due to an error in the cycle length calculation because of large amplitude T waves or low-voltage QRS complexes.
The QT of the 50 patients with a "relatively" short QTc (350 to 360 ms) as reported by the resting ECG system were compared to the results of physician measurements. The resting ECG system reported a mean QTc of 355 ± 3 ms (range, 350 to 360 ms), whereas the physician manually calculated a mean QT of 366 ± 13 ms (range, 353 to 410 ms; p = 0.00023). The change between the computer and physician measurements was 40 ms in 45 of the 50 patients, and between 41 and 80 ms in the remaining patients. In this limited group of short QTc measurements, the resting ECG system indeed did not fail to detect ECGs with a QTc of < 300 ms because of a tendency to overestimate QT.
There were 80 women and 70 men among the 150 patients selected with normal ECGs to learn the distribution of QT intervals. The mean QTc for women was 436 ± 19 ms, and for men was 425 ± 20 ms. Combining sexes, the 95% confidence interval for QTc overall was 392 to 468 ms (mean QTc, 430 ± 19 ms). This is represented in Figure 1 . Assuming a normal distribution, a QTc of < 300 ms would then reflect > 5 SDs. In a normal population of 106,452 patients, 5 SDs from our mean QTc of 430 ms would represent < 1 patient. Presumably, hospitalized patients would tend to have longer QTc intervals than healthy subjects because of, for example, disease or drugs. However, this may not shift the entire distribution to longer QTc intervals but rather might skew the data to longer QTc intervals. We indeed found a negatively skewed distribution, but the kurtosis is that of a normal distribution, not a bimodal one (Fig 1).
|
QT prolongation with attendant arrhythmias has been well described, but only in the last few years have we been alerted to look for short QT intervals because of their association with sudden death.2 However, the definition of a short QT interval appears unclear. In their description of QT intervals, Lepeschkin and Surawicz11 delineated the limits of the normal QT for each heart rate as ± 10% of the average. Using the Bazett formula, the lower limit of the QTc would therefore be 350 ms. A similar lower limit of the normal QTc was confirmed by Vincent et al12 from an analysis of long QT carriers compared to "noncarriers." For the outpatient group of noncarriers, the mean QTc was 420 ms with a range of 380 to 470 ms (n = 116).12 From a larger group (n = 581) examined by Moss,13 the mean QTc of "healthy" individuals was 420 ms (range, 370 to 480 ms). Our hospital population, based on our sample of 150 normal ECG findings had a mean QTc of 430 ms with a (95% confidence interval, 392 to 468 ms).
Short QT syndrome, as characterized by an extremely short QTc, is apparently quite rare. This does not obviate an association between short QTc and arrhythmia or sudden death. There are emerging data that a spectrum of short QT intervals is associated with arrhythmia. Algra et al14 analyzed 6,693 patients undergoing 24-h Holter monitoring. At the 2-year follow-up, the relative risk for sudden death in the setting of a prolonged QT interval was 2.3, not unexpected. Surprisingly, a short QTc (< 400 ms) had an even higher risk. Recently, there have been two reports of relatively short QTc intervals and apparent sudden death. Viskin15 in a series of 28 patients with "idiopathic ventricular fibrillation," suggested that a QTc of < 360 ms may entail increased arrhythmic risk. Maury et al16 described a patient resuscitated from ventricular fibrillation with a subsequent short QTc (315 to 375 ms). Both of these reports raise the possibility that a "relatively" short QTc could increase the risk of arrhythmia.
Based on a genetic analysis of affected families, a short QT interval appears to be the deadliest channelopathy to date.3 Whereas a prolonged QT interval suggests a risk of torsades de pointes, a shortened QT interval points instead to ventricular fibrillation. There have been > 50 publications regarding the genetics and treatment of this new and dangerous syndrome, but its prevalence remains uncertain. Currently, the European registry comprises < 50 patients.17 In this registry, the upper limits for short QT interval was < 320 ms, and for QTc < 340 ms. Given the possibility that a relatively short QTc increases the risk of arrhythmia, perhaps it would be better to liberalize the definition of this syndrome.
In summary, using the current definition of short QT syndrome (QTc 300 ms),9 we were unable to find a single case in a large general hospital population. This is consistent with a recent review18 of 12,012 healthy subjects in which no ECG with a QTc of < 335 ms was found. Second, one cannot rely on a QTc reported by an ECG computer (there seems to be better correlation between manual and automatic measurement with ECGs showing longer QTcs, whereas an apparently short QTc needs to be manually calculated given the degree of errors found in our review). Finally, the evaluation of the risk of arrhythmia with a relatively short QTc deserves further study.
Study Limitations
There are a few limitations to our study. First, our population included only hospitalized patients who had an indication for an ECG to be performed; therefore, this introduces a bias to the population studied. Second, ECGs of patients with a heart rate of > 100 beats/min were included in the total population, the Bazett formula is inaccurate at heart rate extremes, resulting in overcorrection at high heart rates and undercorrection at low heart rates.19
Footnotes
The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.
Received for publication September 3, 2006. Accepted for publication March 6, 2007.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
