Journal of Parenteral and Enteral Nutrition http://pen.sagepub.com/

Incidence of arrhythmia with central venous catheter insertion and exchange RK Stuart, SA Shikora, P Akerman, JA Lowell, JK Baxter, C Apovian, C Champagne, A Jennings, M Keane-Ellison and BR Bistrian JPEN J Parenter Enteral Nutr 1990 14: 152 DOI: 10.1177/0148607190014002152 The online version of this article can be found at: http://pen.sagepub.com/content/14/2/152

Published by: http://www.sagepublications.com

On behalf of:

The American Society for Parenteral & Enteral Nutrition

Additional services and information for Journal of Parenteral and Enteral Nutrition can be found at: Email Alerts: http://pen.sagepub.com/cgi/alerts Subscriptions: http://pen.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav

>> Version of Record - Mar 1, 1990 What is This?

Downloaded from pen.sagepub.com at UNIV OF UTAH on November 24, 2014

0148-6071/90/1403-0152$02.00/0

JOURNAL OF P A R E N T E R A L A N D ENTERNAL NUTRITION Copyright 0 1990 by the American Society for Parenteral and Enteral Nutrition

Vol. 14, No. 2 Printed in U . S . A .

Incidence of Arrhythmia with Central Venous Catheter Insertion and Exchange REGINAK. STUART,M.D., SCOTT A. SHIKORA,M.D., PAULAKERMAN, M.D., JEFFREY A. LOWELL,M.D., JOHN K. BAXTERM.D., CAROLINEAPOVIAN,M.D., CHARLOTTECHAMPAGNE, R.N., ANN JENNINGS,R.N., MARYKEANE-ELLISON, R.N., AND BRUCER. BISTRIAN,M.D., PH.D. From the New England Deaconess Hospital, Harvard Medical School, Boston, Massachusetts

ABSTRACT. The risk of complication during the insertion or exchange of central venous catheters has been well documented. The majority of complications involve mechanical problems associated with insertion. Although cardiac arrhythmia has been acknowledged as a possible complication, its

incidence has never been quantified. We performed cardiac monitoring on patients during 51 central venous catheter insertions or exchanges to determine the incidence of cardiac arrhythmias during guidewire insertion. Forty-one percent of procedures resulted in atrial arrhythmias and 25% produced some degree of ventricular ectopy, 30% of these were ventricular couplets or greater. Ventricular ectopy was significantly more Insertion and exchange of central venous catheters is a common procedure in most hospitals. At the New England Deaconess Hospital (489 beds) we performed 1440 of these procedures in 1988. T h e possible complications of this procedure have been well documented and include hemothorax, pneumothorax, hemopericardium, air embolism, thromboembolism, a n d arterial and venous Although cardiac arrhythmia is sometimes recognized as a potential complication, its actual incidence has previously not been reported. This study documents the incidence of cardiac arrhythmia during central venous catheter insertion or exchange in a stable population of hospitalized patients receiving total parenteral nutrition ( T P N ) . METHODS

The study subjects were medical or surgical inpatients receiving T P N under the care of the Nutritional Support Service (NSS) a t the New England Deaconess Hospital, Boston, MA. Catheters were inserted or exchanged under the guidelines used by the N S S including suspected catheter sepsis, catheter malfunction, and change in access need. Forty-nine consecutive insertions or exchanges were studied. Two procedures initially planned as exchanges eventually required new insertions (one due to purulent discharge from the line track a s the catheter was withdrawn and the other due t o inadvertent loss of access). Thus a total of 51 exchanges or insertions constitute the data base for this report. Forty-five procedures were performed on standard care Received for publication, March 28, 1989. Accepted for publication, June 26, 1989. Reprint requests: Bruce R. Bistrian, MD, PhD, New England Deaconess Hospital, 185 Pilgrim Rd, Boston, MA 02215.

common in shorter patients (160 & 8 us 168 f 11cm, p < 0.05) and when the catheter was inserted from the right subclavian position (43% ventricular ectopy us 10% at the other sites). Other variables such as age, cardiac history, serum potassium, type of procedure, and catheter brand were not significant. It is our conclusion that over-insertion of the wire causes this cardiac stimulation. Despite the absence of morbidity or mortality in this study, this incidence of ventricular ectopy indicates that there is a distinct possibility of a malignant arrhythmia being precipitated by a guidewire. Some modification of the current protocol for these procedures seems indicated. (Journal of Parenteral and Enteral Nutrition 14:152-155,1990)

floors where cardiac monitoring is not routinely available a n d four in intermediate care units where telemetry is routine. No intensive care patients were included. For the purpose of this study, a portable single lead cardiac monitor with hard copy print-out was utilized. An observer, not performing the procedure, conducted the cardiac monitoring throughout the procedure. A baseline rhythm strip was obtained for comparison and a paper print-out was continued during the entire period of wire insertion. T h e exact time of wire insertion and withdrawal were noted on the print-out. The tracing was later reviewed by a t least two of the authors. All arrhythmias reported represent new ectopy in comparison with the baseline rhythm strip. T h e estimated length of guidewire remaining outside the patient a t the point of maximal insertion was recorded by the observer. If a t any time during the procedure a dangerous, sustained arrhythmia was noted, the observer notified the operator to reposition the wire. All procedures were performed by a hyperalimentation fellow or a senior medical resident under the direct supervision of a fellow. Prior experience with this procedure was estimated t o be 30 to 40 exchanges or insertions for each of the two fellows who had completed a medical residency and 200 exchanges or insertions for the four fellows who had each completed 3 years of surgical training. Assistance was provided by specially trained NSS nurses. All procedures were performed under strict aseptic conditions with 1%lidocaine for local anesthesia. Insertion was performed by localizing the appropriate central vein by aspiration of blood with a 22-gauge needle. An 18-gauge needle was then passed along the same pathway into the vein, the wire was inserted through the needle, needle removed, and the catheter was passed over the

152 Downloaded from pen.sagepub.com at UNIV OF UTAH on November 24, 2014

March/Aprill990

ARRHYTHMIAS W I T H CENTRAL VENOUS CATHETERS

wire into the correct position in the vein. Exchanges were performed by passing a wire into the old catheter, removing the catheter while leaving the wire in place, and then inserting a new catheter by passing it over the wire to the proper position. Two brands of double- and triplelumen catheters were used throughout the study. Arrow catheter kits (Arrow International Inc., Reading, PA) consisted of 35-cm catheters accompanied by J-shaped wires 60 cm in length. T h e Edwards brand kits (Baxter Healthcare Corporation, Edwards Critical-Care Division, Irvine, CA) contained 43-cm catheters and 68.5-cm Jshaped wires. Single-lumen catheters of unknown brands which had been placed a t other hospitals were removed on two occasions. Each operator estimated the amount of wire necessary t o assure proper placement of t h e catheter within the superior vena cava (SVC) and attempted t o insert no more than this amount of wire. Chest x-rays were obtained for new insertions only. Routine chest x-rays after guidewire exchanges were not deemed necessary as proper placement in the SVC is ensured by prior catheter position. Patient characteristics t h a t might predispose to arrhythmia, including age, sex, height, weight, serum potassium, cardiac history, cardiac medications, and diagnosis were collected for each procedure. Results are presented as means f SD. Student's t-test was used for comparison of mean values between the group of patients. Chi-square analysis was used for nonparametric comparisons. A p value less t h a n 0.05 was considered significant.

W

K =) 0

symptomatic except one person who reported palpitations after wire removal. S V T (which had not been present during the exchange) was noted on the monitor and resolved spontaneously. Furthermore, no patient with cardiac ectopy during wire insertion reported any symptoms. Neither patient nor operator would have been aware of the reported arrhythmias had cardiac monitoring not been instituted for the purpose of this study. Variables which correlated with a n increased incidence of ventricular ectopy included patient height and anatomical position of the catheter. T h e mean height of patients with ventricular ectopy was significantly less than t h a t of patients without arrhythmia or with atrial ectopy (160 k 8 us 168 k 11 cm, p < 0.05) (Fig. 3). T h e right subclavian position also predisposed t o ventricular ectopy with 43% of procedures in this position resulting in ventricular arrythmias compared with 10% of procedures a t other sites ( p < 0.001). T h e distance from the right subclavian position t o the right heart is shorter than the distance from any other insertion position. Thus, the two variables which reflect the distance from the insertion site t o the heart correlate with frequency of ventricular ectopy. Variables which demonstrated no significant relationship t o the incidence of ventricular ectopy included the operator performing the procedure, age of the patient, cardiac history, serum potassium (range 3.2-5.8), and type of procedure (new insertion us a n exchange over a guidewire). Catheter brand was also not related t o cardiac ectopy which was surprising since the Edwards catheter contained a 68.5-cm guidewire while the Arrow brand catheters were accompanied by 60-cm wires. Estimation of the amount of wire remaining outside of the patient (and thus a calculation of the amount inserted) also did not correlate with the incidence of ectopy. T h i s is probably a reflection of the dif€iculty in accurately estimating lengths of wire in the range of 50 cm, rather than a true lack of correlation (Table I). Complications for this series of patients were minimal and included: loss of venous access during two guidewire exchanges requiring new insertions, and one subclavian artery puncture with no clinical sequela. Other technical problems included failure t o thread a wire through a severely kinked catheter requiring a new insertion, a n d inability to thread a wire into the SVC from the left subclavian or left internal jugular veins in a patient who had undergone multiple central venous catheterizations in the past and had suspected innominate vein thrombosis. There were no clinically significant complications.

60-

W

DISCUSSION

0

gn LL

153

40

-

0

NEW INSERTION

WIRE EXCHANGE

FIG. 1. Type of central venous access procedure performed.

T h e incidence of ventricular arrhythmias during pulmonary artery catheterization has been reported to be . ~Morbidity from this event is low, between 12 a n d 7 7 % 1 ' largely because continuous cardiac monitoring is required and the catheter can be quickly repositioned when ectopy occurs. There have been no published reports of the incidence of cardiac ectoDv with central line insertion. Ideallv. insertion of a l i i e into the SVC should not entail e;: trance into the right heart by the guidewire or the cath-

Downloaded from pen.sagepub.com at UNIV OF UTAH on November 24, 2014

154

Vol. 14, No. 2

STUART ET AL

1 40

Vent. Tach.

W k

0

U

a

Triplet's Couplet's

d

Total PVC's > 4

Total pvC's

r

a a 4:

FIG.2. Types of arrhythmias during 51 central venous catheter procedures.

10

0

T-

.-m

E 5

m .-

.-m

f

5 %

E

E

%

%

f

f

f

z

z

z

0

z

170

I

T

7

-

-

E

S.D.=l 1.O

S.D.=8 .o 160

L-

P< 0.05 F

z

E

150

4

n

140

-I

Ventricular

None or Atrial

CATEGORY OF ARRHYTHMIA

FIG. 3. Relationship of patient height to incidence of ventricular arrhythmia.

TABLE I Variables correlated with incidence of ventricular ectoDv Significant

Not significant

Patient height Anatomic position

Operator Age Cardiac history Type of procedure Catheter brand

eter; thus cardiac ectopy should never result. However, occasional cardiac monitoring during these procedures as well as anecdotal reports have demonstrated that cardiac ectopy does indeed occur. This study demonstrates a 25% incidence of ventricular ectopy and an 11%incidence of couplets or greater

in the hands of experienced operators. The correlations with the height of the patient and the right subclavian position suggest that over-insertion of the wire might be the precipitant of this ectopy. Critical examination of our data and our technique has led to the conclusion that there are two factors contributing to over-insertion of the guidewire. First, the generous lengths of the double- and triple-lumen catheters (35-43 cm) and the accompanying guidewires (60-68.5 cm) lead to difficulty in accurately estimating the amount of wire which has actually been inserted. Second, during catheter manipulation, both in and out over the guidewire, the wire is frequently moved despite the operator's belief that the wire is being held firmly in place. This often results in further advancement of the wire intravascularly. We found that these errors occur even in simulated clinical scenarios. Certainly in actual clinical situations the combined goals of maintaining venous access, controlling the free end of the wire, preserving sterility, and sometimes calming a tense patient all compete for the operator's attention, making estimation of wire length and catheter manipulation even more challenging. Anecdotal evidence collected over several years at this institution indicates that sustained and dangerous arrhythmias can result from guidewire stimulation. During the last 8 months, since we have become aware of this problem, we have noted two serious incidents during central venous catheter procedures in patients in our hospital. Approximately 950 of these procedures were performed over this period of time. The first incident occurred during a new insertion in a previously stable, unmonitored floor patient. The patient became unresponsive and hypotensive during guidewire insertion. Despite rapid withdrawal of the wire, the patient's status did not spontaneously improve and ventricular fibrillation was noted when cardiac monitoring was established.

Downloaded from pen.sagepub.com at UNIV OF UTAH on November 24, 2014

MarchlAprill990

ARRHYTHMIAS WITH CENTRAL VENOUS CATHETERS

The patient was resuscitated but severe cerebral anoxia resulted. A second event occurred in another unmonitored patient who was undergoing a catheter exchange. The patient became syncopal and hypotensive during guidewire insertion. The clinical deterioration responded to guidewire withdrawal and intravenous fluids. Cardiac monitoring was not established until after the patient had recovered and revealed normal sinus rhythm at that time. Recovery was uneventful. Both procedures were performed by senior surgical residents with extensive experience with central venous catheterization. During neither of these events was there any indication of air embolism. Although there was no definite evidence that an arrhythmia was the cause of either of these events, it was highly suspected in both cases. There was no morbidity, mortality, or even symptoms during the 51 procedures in this study, but there was a 25% incidence of ventricular ectopy and a 4% incidence of ventricular tachycardia. Over the course of 1 year at this institution this rate would translate into approximately 57 patients being stimulated into ventricular tachycardia by guidewires. If any one of these patients had an irritable myocardium or if the wire remained in place for more than a brief period of time, sustained ventricular tachycardia could result. Thus, our results seem to confirm the anecdotal reports regarding this rare, but life-threatening, complication of central venous catheter placement. Some modification of the current procedure protocol seems warranted to eliminate, or greatly reduce, this risk. If, as our data indicates, over-insertion of the guidewire is the causative factor, there is reason to believe that the incidence of this event may have increased in recent years. Five years ago the NSS at this institution was using single-lumen, 15-cm catheters and 35-cm guidewires. Today we typically utilize catheter kits containing double- or triple-lumen 43-cm catheters with 68.5-cm guidewires. The potential for over-insertion of the guidewire or the catheter would appear to be greater with the current equipment. Thus the number of serious cardiac arrhythmias precipitated by central line insertion may actually have increased over the past few years despite more familiarity with this technique. This may explain

155

why this complication has not been appreciated in the past. Since stable floor patients such as those included in this study generally undergo central line placement without cardiac monitoring, our results raise the question of whether cardiac monitoring should be mandatory during these procedures. This would, of course, have serious cost implications given the need for a monitor and additional personnel. It is the authors’ conclusion that modification of the present technique or equipment, with the goal of eliminating guidewire over-insertion, would result in significantly fewer ventricular arrhythmias and obviate the need for cardiac monitoring. Further study of this issue is obviously necessary in light of the serious health and economic implications. ACKNOWLEDGMENTS

The authors gratefully acknowledge the assistance of Timothy Babineau, MD for the preliminary data that inspired this study and of Eugene F. Foley, MD for his expert assistance with computerized data analysis. REFERENCES 1. Padberg FT, Ruggiero BS, Blackburn GL, et al: Central venous catheterization for parenteral nutrition. Ann Surg 193:264-270,

1981 2. Sitzman JV, Townsend TR, Siler MC, e t al: Septic and technical complications of central venous catheterization. Ann Surg 202:766770, 1985 3. Bernard RW, Stahl WM: Subclavian vein catheterizations: A prospective study. Ann Surg 173:184-190, 1971 4. Feliciano DV, Mattox KL, Graham JM, et al: Major complications of percutaneous subclavian vein catheters. Am J Surg 138:869-874, 1979 5. Damen J : Ventricular arrhythmias during insertion and removal of pulmonary artery catheters. Chest 88:190-193, 1985 6. Elliott CG, Zimmerman GA, Clemmer TP: Complications of pulmonary artery catheterization in the case of critically ill patients. Chest 76:647-652, 1979 7. Iberti TJ, Benjamin E, Gruppi L, et al: Ventricular arrhythmias during pulmonary artery catheterization in the intensive care unit. Am J Med 78:451-454, 1985 8. Sprung CL, Pozen RG, Rozanski JJ, et al: Advanced ventricular arrhythmias during bedside pulmonary artery catheterization. Am J Med 72:203-208, 1982

Downloaded from pen.sagepub.com at UNIV OF UTAH on November 24, 2014