a b s t r a c t

Introduction: Prolongation of the QT interval is a well-recognized complication associated with many commonly used medications. Emergency Department monitoring of the corrected QT (QTc) both before and after medica- tion administration is typically performed using the 12 lead electrocardiogram (ECG). The purpose of this study is to compare the QTc reported on the 12 lead ECG to that reported by single brand of bedside monitor. Methods: A convenience sample of emergency department patients over the age of 18 undergoing bedside mon- itoring and who had an ECG ordered by their treating physician were enrolled. These patients underwent simul- taneous ECG and monitor QTc calculation. The primary outcome of interest was the correlation between the monitor and ECG QTc. Secondary outcomes included ability of each method to identify patients with a QTc N 500 ms and the ability of each method to identify patients with a QTc b 450 ms.

Results: A total of 125 patients had simultaneous ECG and monitor QTc measurements recorded. There was mod- erate correlation between the monitor and ECG QTc (Pearson’s correlation coefficient = 0.55). The median dif- ference between the ECG QTc and the monitor QTc (ECG QTc minus monitor QTc) was -7 ms (IQR -23 to 11 ms).

Conclusion: We found that there was moderate correlation between the QTc reported on the 12 lead ECG and that reported by the bedside monitor. This correlation is not strong enough to support the use of the bedside monitor as a substitute for the 12 lead ECG when evaluating a patient’s QTc.

(C) 2017

Introduction

Drug induced QT interval prolongation is a well-recognized compli- cation of many commonly prescribed pharmaceuticals from a variety of drug classes. Prolongation of the QT interval is associated with Torsades de pointes(TdP), a potentially lethal cardiac dysrhythmia. This rare dys- rhythmia was first described in 1966 [1] and was found to be the under- lying cause of “quinidine syncope”, a side effect of quinidine therapy that was first recognized in the 1920s [2,3]. The pathophysiology of TdP is complex and not fully predicted by the QT alone; however, the QT measurement is the simplest and most rapid screening tool used to identify patients at risk of TdP. The risk of TdP remains a limiting factor in the utilization of many cardiac medications, however this adverse ef- fect is not limited to this drug class [3,4]. In fact, drug induced QT inter- val prolongation has resulted in the restriction or complete removal

* Corresponding author at: University of California - Davis, Department of Emergency Medicine, Sacramento, CA, United States.

E-mail address: [email protected] (J.A. Chenoweth).

from the market of several unrelated non-cardiac medications including cisapride, grepafloxacin, and droperidol [3].

With increased focus on the risks associated with drug induced QT prolongation come new challenges for clinicians treating patients in the acute care setting. The list of medications that have been associated with QT prolongation and risk of TdP is constantly growing. Review of the website crediblemeds.org, an independent online database man- aged by the Arizona Center for Education and Research on Therapeutics, reveals a total of 142 drugs that can cause QT prolongation [5]. Due to these risks many hospitals have implemented protocols requiring docu- mentation of a rate corrected QT (QTc) prior to medication administra- tion. This rate correction most commonly uses Bazett’s formula [6], although alternative formulas have been proposed which do not overes- timate the QTc in the presence of tachycardia [7].

Currently, the primary method used to assess the risk of TdP in most emergency departments is the calculated QTc on a 12 lead electrocar- diogram (ECG). While accurate, this method is cumbersome and not practical for real time monitoring during treatment. One alternative is the QTc monitoring readout on the bedside monitor. Using a patient’s bedside monitor would allow for real-time QTc monitoring during and

https://doi.org/10.1016/j.ajem.2017.10.012

0735-6757/(C) 2017

778 J.A. Chenoweth et al. / American Journal of Emergency Medicine 36 (2018) 777-779

after the administration of medications that are known to prolong the QTc. The purpose of our study is to compare the QTc reported by the bedside monitor to that calculated from a standard 12 lead ECG. We hy- pothesize that the bedside monitor is an accurate and simple method for QTc monitoring.

Methods

This is a prospective cross-sectional study performed at a single aca- demic urban emergency department between 8/28/2013 and 6/11/ 2014. A convenience sample of patients 18 and older undergoing cardi- ac monitoring (Philips(C) Intellivue MP50) and who had an ECG (GE(C) Healthcare MAC 5500) ordered by their treating physician were en- rolled in the study. This study was approved by the local institutional re- view board with a waiver of patient consent.

After enrollment, a simultaneous ECG and cardiac monitored rhythm strip were obtained. Demographic data including age, electro- cardiogram intervals as reported by the monitor or ECG machine, heart rate and sex were recorded for analysis. The primary outcome of interest was the correlation between the monitor and ECG QTc. Second- ary outcomes included ability of each method to identify patients with a QTc N 500 ms, the ability of each method to identify patients with a QTc b 450 ms, and the effect of heart rate on the agreement between the monitor and a 12 lead ECG. The N 500 ms cutoff was chosen because of the known association between a QTc N 500 ms and risk of TdP. The

<= 450 ms cutoff was chosen because this is the cutoff used at our facility for intravenous administration of certain QT prolonging medications.

For the purpose of this study we used the QTc instead of the numeric QT to factor in correction for patient heart rate. While there is no univer- sal agreement of which formula best corrects for heart rate, we chose to use the machines’ (monitor and ECG) calculation both of which use Bazett’s formula for heart rate correction. We felt the use of the machine calculated QTc would most accurately reflect clinical practice.

Once collected, all data were entered into an Excel(C) spreadsheet

(Microsoft Corp, Redmond, WA, USA). The data were analyzed by the use of STATA 14.2 (StataCorp LLC, College Station, TX, USA). Data nor- mality was tested with the Shapiro-Wilk test with normally distributed data reported as means with standard deviations and non-parametric data reported as medians with interquartile ranges. A Pearson’s correla- tion coefficient was calculated comparing the monitor and ECG QTc with a subgroup analysis performed based on gender, for tachycardic patients, and patients with normal heart rates. There were insufficient bradycardic patients available to perform an accurate comparison.

Results

A total of 135 patients were evaluated for inclusion in the study. Ten patients were missing monitor QTc measurements leaving 125 with si- multaneous ECG and monitor QTc intervals recorded. Demographic and basic statistical data can be seen in Table 1. Mean age was 56.7 years (SD

18.7 years). There were 88 males (70.4%) and 37 females (29.6%) in- cluded in the study. The median difference between the ECG QTc and the monitor QTc (ECG QTc minus Monitor QTc) was -7 ms (IQR -23 to 11 ms).

A total of 30 patients had either an ECG or monitor QTc N 500 ms (10 monitor only, 7 ECG only, 13 both). The monitor identified more of the patients with a QTc N 500 ms (76.7%, 95% CI 57.7-90.1%) than were iden- tified on the ECG (66.7%, 95% CI 47.2-82.7%). There were 81 patients with a QTc <= 450 ms (17 monitor only, 19 ECG only, 45 both). The ECG would allow IV administration of QT prolonging medication in 51.2% of patients (95% CI 42.1-60.2%) while the monitor would allow QT pro- long drug administration in 49.6% of patients (95% CI 40.5-58.7%).

A comparison of monitor and ECG QTc can be seen in Fig. 1. Overall the monitor and ECG had moderate correlation (r = 0.55). There was better agreement between the monitor and ECG QTc in female patients compared to males (r = 0.63 vs 0.50). When evaluated in groups based on heart rate, there was slightly better agreement between the monitor and ECG QTc for patients that had normal heart rates than those that were tachycardic (r = 0.55 vs 0.54).

Discussion

In this study we found a moderate correlation between the monitor and ECG QTc measurements. This was true in both tachycardic patients and those with normal heart rates. Interestingly, even though the mon- itor QTc tended to be less than that on the ECG, the monitor identified more patients with a QTc N 500 ms. It was also noted that the monitor QTc would have been more restrictive when it comes to the administra- tion of QTc prolonging medications by identifying fewer patients as hav- ing a QTc <= 450 ms. However, it remains unclear which method is more accurate. Given this uncertainty there is insufficient evidence to support the use of the bedside monitor as a substitute for the gold standard, the 12 lead ECG. This is particularly true when making decisions regarding the administration of medications known to prolong the QTc.

Many methods to evaluate the risk of developing TdP have been de- veloped. These include the QT nomogram which incorporates the un- corrected QT in relation to the heart rate, the “Half RR rule”, and standard QTc using either Bazett’s or Fridericia’s correction equations. One study by Berling and Isbister [8] compared each of these methods using a previously published data set of TdP cases and controls. In this study the QT nomogram had a sensitivity of 96.9% and a specificity of 98.7%. When comparing the two QT correction equations, Bazett’s was more sensitive than Fridericia’s (93.8% vs 82.2%) while Fridericia’s was more specific (100% vs 97.2%). The “Half RR rule”, which is the basis for most “eyeball” assessments for QT prolongation, had a sensitivity of 87.6% and a specificity of 52.9% [8]. This study highlights the

Table 1 Baseline patient characteristics.
	Male (n = 88)	Female (n = 37)
Age (years)	54.8 (18.4)a	61.3 (19.1)a
QTc (ms) Monitor	447 (426-483)b	468 (447-490)b
ECG	449 (428-479)b	452 (433-486)b
QTc N 500 (ms) Monitor	17.0% (9.9-26.6%)c	21.6% (9.8-38.2%)c
ECG	17.0% (9.9-26.6%)c	13.5% (4.5-28.8%)c
QTc <= 450 (ms) Monitor	56.8% (45.8-67.3%)c	32.4% (18-49.8%)c
ECG	53.4% (42.5-64.1%)c	45.9% (29.5-63.1%)c

^a Mean (standard deviation).

^b Median (interquartile range).

^c 95% confidence interval. Fig. 1. Correlation between ECG and monitor QTc.

J.A. Chenoweth et al. / American Journal of Emergency Medicine 36 (2018) 777-779 779

importance of actual QT measurements when evaluating risk of devel- oping TdP.

The ability to identify patients at increased risk for TdP prior to ad- ministration of QT prolonging medication has important implications for patient safety. One study of patients admitted to a cardiac care unit found that 34.7% of patients with prolonged QTc on admission were still administered QT prolonging medications during their hospitaliza- tion. Even more concerning, 42.2% of patients with an admission QTc N 500 ms received QT prolonging medications during their hospitaliza- tion [9]. The authors identified QTc documentation as one of the prob- lems, with only 60% of patients having QTc documentation within 24 h of receiving a QT prolonging medication [9]. This conclusion is con- sistent with a study from the Veteran’s Affairs Pittsburgh Healthcare System which found that only 19% of providers actually documented the presence of a QTC N 500 ms [10].

The problem of documentation was also identified in a study by Sandau et al. The authors found that only 17.3% of inpatients on contin- uous ECG monitoring had appropriate QTc documentation in the elec- tronic medical record (EMR) prior to receiving QT prolonging medications. Using a combination of a teaching intervention of best practice advisories in the EMR they were able to increase this to 62.1% [11]. This study highlights the potential utility of EMR systems for in- creasing QTc documentation and improving patient safety.

Advances in EMR and monitor integration have the potential to fur- ther improve patient safety. Currently, many systems are able to directly transfer vital signs into the EMR. If this process was used for the QTc as well, it would be possible to incorporate an alert that would show up prior to the administration of QT prolongation medications only in pa- tients with current QTc prolongation. Such alerts could either guide pro- viders to use alternate medications or take actions such as electrolyte repletion aimed at mitigating the risk of QT prolongation and TdP.

Limitations

This study has several limitations. First, only automated calculations of QTc were used and no manual over-read was performed. This could result in calculation errors due to patient movement, particularly for the QTc on the monitor. Additionally, there were likely differences in lead placement between the monitor and ECG. The monitor ECG only uses 5 leads from which it is able to calculate 7 different leads. Differ- ences in lead placement could result in differences in the calculated QTc, particularly in patients with significant QTc dispersion. However, both of these situations reflect current clinical practice. Finally, this

was a relatively small study focused on a single academic emergency department. There is significant QTc variability between patient popula- tions which could limit the generalizability of our findings.

Conclusions

We found that there is moderate correlation between the monitor and ECG QTc. In our patient population, the monitor was able to identify more patients with a QTc N 500 ms; however, no manual corroboration of this finding was performed. The bedside monitor has multiple advan- tages over 12 lead ECGs for QTc monitoring including ability to auto- matically transfer data to the EMR and provide continuous readings. However, our results do now show a strong enough correlation to sup- port the conclusion that the bedside monitor can be used instead of a 12 lead ECG, particularly when making Clinical decisions regarding medica- tion administration. More data is needed to confirm the accuracy of the monitor QTc compared to a standard 12 lead ECG.

References

1. Dessertenne F. Ventricular tachycardia with 2 variable opposing foci. Arch Mal Coeur

   Vaiss 1966;59:263-72.

   Selzer A, Wray HW. Quinidine syncope. Paroxysmal ventricular fibrillation occurring during treatment of chronic Atrial arrhythmias. Circulation 1964;30:17-26.

2. Roden DM. Drug-induced prolongation of the QT interval. J Med]->N Engl J Med 2004;350:1013-22.
3. Trinkley KE, Page 2nd RL, Lien H, Yamanouye K, Tisdale JE. QT interval prolongation and the risk of torsades de pointes: essentials for clinicians. Curr Med Res Opin 2013; 29:1719-26.
4. Woosley RaR KA. QT drugs list. www.crediblemeds.org, Accessed date: 27 February 2017.
5. Bazett H. An analysis of the time-relations of electrocardiography. Heart 1920;7: 353-70.
6. Chiladakis J, Kalogeropoulos A, Arvanitis P, Koutsogiannis N, Zagli F, Alexopoulos D. Preferred QT correction formula for the assessment of drug-induced QT interval pro- longation. J Cardiovasc Electrophysiol 2010;21:905-13.
7. Berling I, Isbister GK. The half RR rule: a poor rule of thumb and not a risk assess- ment tool for QT interval prolongation. Acad Emerg Med 2015;22:1139-44.
8. Tisdale JE, Wroblewski HA, Overholser BR, Kingery JR, Trujillo TN, Kovacs RJ. Preva- lence of QT interval prolongation in patients admitted to cardiac care units and fre- quency of subsequent administration of QT interval-prolonging drugs: a prospective, observational study in a large Urban academic medical center in the US. Drug Saf 2012;35:459-70.
9. Good MM, Riad FS, Good CB, Shalaby AA. Provider response to QTc prolongation on standard 12-lead EKG: do we notice or do we care? Pacing Clin Electrophysiol 2016; 39:1174-80.
10. Sandau KE, Sendelbach S, Fletcher L, Frederickson J, Drew BJ, Funk M. Computer- assisted interventions to improve QTc documentation in patients receiving QT- prolonging drugs. J Crit Care]->Am J Crit Care 2015;24:e6-15.