- Email: [email protected]
Bedside Chest Radiographs and How Ambiguous Peripherally Inserted Central Catheter Tips Happen: A Case Report
C A S E
R E P O R T
Bedside Chest Radiographs and How Ambiguous Peripherally Inserted Central Catheter Tips Happen: A Case Report Stuart Gordon, MSN, RN, CRNI, VA-BC, RN-BC Legacy Emanuel, Portland, OR
Abstract Verification of peripherally inserted central catheter (PICC) tip location with radiographic imaging is at times challenging. This case report details the experience of 1 patient and the difficulties encountered identifying the PICC tip using bedside chest radiographs. It examines, offers to explain the issues, and reviews the results of 209 chest radiographs post-PICC placement for consistency. Keywords: PICC confirmation radiograph, radiograph gold standard, bedside chest radiograph
Introduction he tip of a peripherally inserted central catheter (PICC) should ideally be in the distal one-third of the superior vena cava (SVC) or cavoatrial junction (CAJ), but correctly identifying placement using bedside chest radiographic imaging is often difficult. The following is a detailed examination of the errors introduced with bedside chest radiographs, followed by a case report. Inaccuracies seen with bedside chest radiographs may lead to unnecessary manipulation or replacement of a PICC; may increase costs to patients; may increase the risk of infection; and may result in the PICC tip terminating, unseen, in a location more distal than the CAJ. The purpose of this article is to bring into focus how bedside chest radiographs fail. The use of bedside radiographic imaging to verify central line tip location and identify placement-related complications is considered by many to be the only acceptable assessment method postinsertion. However, those placing PICC lines have experienced first-hand the difficulties of seeing the tip or accurately judging the appropriate distance to withdraw the PICC when the tip is determined to be in the atria. Members of vascular access teams placing lines are more intimately familiar with the limitations and frustrations than the independent providers who only see the results of a team’s
T
Correspondence concerning this article should be addressed to [email protected] http://dx.doi.org/10.1016/j.java.2016.08.001 Copyright © 2016 ASSOCIATION FOR VASCULAR ACCESS. Published by Elsevier Inc. All rights reserved.
work. With the widespread introduction of electrocardiogram (ECG), ECG/Doppler, and ECG/magnetic PICC tip locators the limitations of radiographs have only been magnified. Accurately identifying the location of a PICC tip with initial placement saves nursing time and patient cost. If, for example, the tip of a PICC is seen projecting below the CAJ institutional policy may require retraction of the line to bring it out of the atria. This requires nursing time to perform the adjustment, it delays use of the line, it requires disruption of the overlying dressing (an intervention that has been shown to increase the risk of a central lineassociated bloodstream infection1), and it will increase the overall cost to patients. Alternatively, if the tip of a PICC is seen at the confluence of the brachiocephalic veins many will remove and replace the PICC, again increasing cost, nursing time, and the risk of infection. This case study suggests initial bedside chest radiograph errors may lead to unnecessary adjustments of PICC lines. Discussion The impediments to accurately identifying the tip of a PICC with a bedside chest radiograph are patient position in relation to the source of radiographic images, whether the radiograph is taken during inspiration or expiration, cardiac motion blur, interreader variability, patient position, and patient body habitus. For the following examples and for the case study it may be helpful to think of radiographic results as shadows. A radiation source will emit high-energy photons that travel through or are blocked by, in this case, the bones, organs, and a central line in the chest on their way to the radiography plate. Parallax error will occur when the radiographic source and the radiography plate are not in 90 alignment. The idea of
2016
j
Vol 21 No 4
j
JAVA
j
237
Central line X-ray source
here in alignment with actual
X-ray plate
Tip seen
locaƟon C
A
Figure 1. Fluoroscopic placement of the port.
image
after
initial
parallax can be visualized by thinking of a right triangle, where the source of the radiograph is point A, the path of photons from the high energy (radiograph) source is side B, and the resulting image of the tip is point C (Figure 1). If the source of photons is not 90 to the plate the path of photons becomes side C and the image of the tip becomes point B instead of the actual tip at point C (Figure 2). Using the equation p2b a ¼ p 2ðpbÞ (solving for side a of a right triangle) if the source of radiographs is 7 either side of 90 (ie, 83 or 97 ) and the distance from point A to point C is 15 cm (average distance from the tip of the catheter to the radiography plate in a 70-75 kg patient) there will be a 2 cm (rounded up) error in identifying where the tip of the catheter appears. Although 2 cm may seem trivial, if the radiograph shows the PICC 2 cm into the atria due to parallax, the result may be an unnecessary retraction. Having the source of radiographs and the radiography plate in correct alignment (as they are set in a radiology department) would all but eliminate parallax; however, taking a patient to a radiology department is not always an option. The ability to see the tip of a PICC is in part dependent on the size of the patient. Adequate penetration of x-rays through tissue is contingent on exposure to the source and larger patients are exposed for longer times. Longer exposure times increase the chance of having cardiac motion and blood turbulence whip the catheter, blurring the image in the same way long camera exposure times will blur pictures. This is often referred to as cardiac motion blur and as the fluoroscopic images in Figures 3 and 4 show, it is not limited to PICC catheters nor is it limited to bedside chest radiographs.
Actual locaƟon
X-ray plate
Central line
Due to parallax Ɵp is seen here
Figure 2. Fluoroscopic image before port revision secondary to palpitations.
238
j
JAVA
j
Vol 21 No 4
j
Figure 3. Fluoroscopic image of a port. Figures 3 and 4 are of the same patientdFigure 3 after initial placement of the port and Figure 4 before port revision secondary to palpitations. Figure 5 shows a PICC, immediately postplacement, whereas Figure 6 is of the same patient 3 days later. This patient’s electronic health record was reviewed to verify that the line had neither been manipulated nor had it migrated, it was cardiac motion that blurred the distal end of the PICC. Several factors contribute to the differences in interpretation of tip location, the first being interreader variability. Different radiologists will use different landmarks when interpreting what they see, and being able to see the PICC tip isdas noted abovedsometimes difficult. There is little consensus as to how well bedside chest radiographic images correlate to the CAJ when using common landmarks. Vesely states, “A catheter tip positioned 3 cm below the right tracheobronchial angle would always be in the SVC,”2 and in a 2008 study Verhey found, “The average distance from the superior vena cavaeright atrial appendage to the cavoatrial junction was 1.8 cm” on chest radiographs in adult patients.3 A similar conclusion was reached by Baskin et al,4 who found “the true cavoatrial junction is located more inferiorly than commonly believed and is not accurately estimated with commonly used imaging landmarks.” The population in that study was predominately pediatric patients lying in a supine position and the authors reviewed computerized tomography (CT) results. Another study using CT found tip placement 4 cm below the carina “will result in placement near the cavoatrial junction.”5 Perhaps the results of these articles is best summed up by Wirsing et al,6 who used transesophageal echocardiography to assess accuracy of bedside chest radiographs and determined, “Reading of a bedside chest x-ray alone is not very accurate to identify intraatrial [central venous catheter] tip location.” They went on to say, “Although bedside [chest radiograph] can be a convenient screening tool for some problems, it cannot serve as a ‘gold standard’ for assessing a [central venous catheter] tip position with respect to the right atrium.”
2016
Figure 6. Three days after placement. Figure 4. Port visualized with fluoroscopy. The port was revised due to palpitations. Cardiac motion blur is present. Three additional factors influencing PICC tip location are arm position, phase of inspiration/expiration, and whether the patient is lying or standing for the radiograph. Respiratory motion was shown to alter central line tip location by 9 mm in 1 study.7 Adduction of the arm from 90 to 0 will move the tip of the catheter 2 cm on average, an effect that is greater
Figure 5. Bedside chest radiograph of a peripherally inserted central catheter, immediately postplacement.
on the right than the left,8 and moving from a supine position to a standing position will cause abdominal contents to shift in a caudal direction an average of 3.2 cm.9,10 Several articles written in recent years have questioned the accuracy of bedside chest radiographs for central line tip placement when compared to ECG systems,11,12 whereas another article13 questioned whether bedside chest radiographs should be considered the gold standard. Challenging old-guard practices have come after gaining a deeper understanding of the limitations of bedside radiographs and those factors reducing accuracy.14 To gain a better understanding of how often the challenges with bedside chest radiographs occur, a research proposal was submitted to our institution’s institutional review board that granted approval for both a case report and a retrospective chart review. Case Report Consider the case of a 49-year-old, 145-kg patient admitted with a chief complaint of sudden onset severe abdominal pain, subsequently diagnosed with gastric perforation. He had surgical intervention, was intubated, and several days into his hospitalization his central line was removed and a PICC was ordered for long-term antibiotics and total parenteral nutrition administration. After the PICC was successfully placed (using external landmark measurements and without the aid of magnetic, ECG, ECG/magnetic, or ECG/Doppler guidance) a radiographic image was taken and read by a radiologist. It was determined that the right PICC catheter tip was in the SVC approximately 6 cm above the CAJ. Both the nurse inserting the PICC and the ordering physician were unable to see the tip of the PICC with clarity and ordered a repeat radiograph. There had been no manipulation of the line in the interval and the radiologist determined that the distal PICC was in the SVC, tip not well seen. The radiologist, now being unable to see the tip, suggested a third radiograph,
2016
j
Vol 21 No 4
j
JAVA
j
239
in which the right arm PICC was visible to the level of the mid-SVC but the location of the tip was ambiguous. A fourth radiograph was ordered, taken, and it was interpreted that the tip of the right arm PICC was near the CAJ. It is worth noting again that there had been no manipulation of the line, and despite 3 prior radiographs, the single radiograph in which the tip was interpreted to be in the ideal location was the radiograph accepted as accurate, an interesting example of confirmation bias, or the tendency to find information that corroborates existing belief. At our institution it is standard for all intubated intensive care unit patients to have daily chest radiographs and for our case report these provide further examples of the challenges with bedside radiographs. Day 1 post-PICC placement it was reported that the central venous catheter was projecting from the right neck with its tip in the plane of the SVC about 4 cm above the CAJ. Day 2 post-PICC placement it was reported that the central venous catheter tip was in the plane of the SVC about 5.3 cm above the CAJ. Day 3 post-PICC it was reported that the right PICC catheter tip was in the SVC at the approximate level of the CAJ. On Day 4 it was reported that the right PICC catheter tip was at the CAJ, and on Day 5 it was reported that the right PICC catheter tip was in the right atrium approximately 2 cm below the CAJ. A quick review of the radiographs showed the tip appeared in different locations, from 6 cm above the CAJ to 2 cm into the atriada total of 8 cm although there had been no manipulation of the line, the patient’s arms were at his side for each radiograph, and the head of the bed was positioned at the same angle. Findings The patient experience noted above is an outlier but illustrates the point that bedside chest radiographs are not infallible. The author reviewed 414 post-PICC placement radiographic images as a routine quality measure. A radiologist interpreted the PICC tip at the CAJ in 110 patients with the tip seen an average of 4.5 cm below the carina, the extremes being 0-10 cm. Only 209 of the 414 patients (50%) had more than 1 radiograph after placement, and these were assessed for consistency in identified tip location, and the electronic health records of each were reviewed to confirm there had been no manipulation or migration of the line that would explain the observed discrepancy. In these serial radiographs the PICC tip was identified in a location that differed from the original in 68 (32%), ranging from 1.5-8 cm. As an example of the extremes in variation, the results of 1 PICC are offered where the tip was visualized “at about the level of the carina, which measures approximately 6.5 cm above the CAJ.” The PICC migrated out 2 cm and to confirm tip location another radiograph was taken after which it was determined that the right PICC catheter tip was in the SVC above the CAJ. Eight days later, with no migration or manipulation of the line, another routine bedside radiograph was taken and read as “the tip is in the SVC approximately 8.5 cm above the CAJ.” Three of the patients above had a CT scan after placement. One of the 3 had 2 radiographs before the CT scan that were
240
j
JAVA
j
Vol 21 No 4
j
reviewed in the interpretation of the second radiograph, which reported that “both images demonstrated the PICC at the level of the carina about 4 cm above the CAJ.” It was noted that the first image was “questionable for the PICC being visible in the right atrium about 1 cm beyond the CAJ.” Two days later the CT scan showed the tip of the central venous catheter was in the lower right atrium. The second patient had the tip of the PICC identified as in the proximal right atrium with radiograph, 4 days later with CT the PICC tip was seen in the proximal SVC. The third patient’s PICC tip was seen on radiograph in the proximal SVC within 1 cm above the CAJ, the next day during the CT scan the tip was seen projecting at the caudal aspect of the right atrium inferior vena cava junction. One patient had a transesophageal echocardiogram (TEE) after PICC placement. The immediate postplacement radiograph showed that the PICC tip was projecting at the CAJ while the PICC was seen extending beyond the tricuspid valve on the TEE. The electronic medical records of the 3 patients with CT scans and the 1 patient with the TEE were reviewed to confirm that there was no manipulation or migration of the PICC. Conclusion As discussed, there is a degree of inaccuracy with bedside chest radiographic imaging and it is hoped that this article opens a discussion about the use of alternative modalities for identifying PICC tip location such as ECG. However, using bedside chest radiographic imaging as a method to gauge accuracy of an ECG device may produce some incongruous results. If alternate methods such as ECG are not an option, having postplacement radiographic images taken in a radiology department when possible will eliminate parallax and may reduce the rate of placements where the tip is not well seen. Disclosures The author has no conflicts of interest to disclose. References 1. Timsit JF, Bouadma L, Ruckly S, et al. Dressing disruption is a major risk factor for catheter-related infections. Crit Care Med. 2012;40(6):1707-1714. 2. Vesely T. Central venous catheter tip position: a continuing controversy. J Vasc Intervent Radiol. 2003;14: 527-534. 3. Verhey P. The Right mediastinal border and central venous anatomy on frontal chest radiography-direct CT correlation. J Assoc Vasc Access. 2008;13(1):32-35. 4. Baskin K, Jimenez R, Cahill A, Jawad A, Towbin R. Cavoatrial junction and central venous anatomy: implications for central venous access tip position. J Vasc Intervent Radiol. 2008;19:359-365. 5. Mahlon M, Yoon H. CT Angiography of the superior vena cava: normative values and implications for central venous catheter position. J Vasc Intervent Radiol. 2007;18: 1106-1110. 6. Wirsing M, Schummer C, Neumann R, Steenbeck J, Schmidt P, Schumer W. Is traditional reading of the
2016
7.
8.
9.
10.
bedside chest radiograph appropriate to detect intraatrial central venous catheter position? Chest. 2008;134(3): 527-533. Pan PP, Engstrom BI, Lungren MP, Seaman DM, Lessne ML, Kim CY. Impact of phase of respiration on central venous catheter tip position. J Vasc Access. 2013;14(4):383-387. Forauer A, Alonzo M. Change in peripherally inserted central catheter tip position with abduction and adduction of the upper extremity. J Vasc Intervent Radiol. 2000;11(10):1315-1318. Nazarian G, Bjarnason H, Dietz C, Bernadas C, Hunter D. Changes in tunneled catheter tip position when a patient is upright. J Vasc Intervent Radiol. 1997;8:437-441. Kowalski C, Kaufman J, Rivitz S, Geller S, Waltman A. Migration of central venous catheters; implications for
11.
12.
13.
14.
2016
initial catheter tip positioning. J Vasc Intervent Radiol. 1997;8:443-447. Gebhard R, Szmuk P, Pivalizza E, Melnikov V, Vogt C, Warters R. The accuracy of electrocardiogram controlled central line placement. Anesthes Analg. 2007;104:65-70. Pittiruti M, Scoppettuolo G, LaGreca A, et al. The EKG method for positioning the tip of PICCs: results from two preliminary studies. J Assoc Vasc Access. 2008;13(4): 112-119. Oliver G, Jones M. ECG or X-ray as the ‘gold standard’ for establishing PICC-tip location? Br J Nurs. 2014;23(19 Suppl):S10-S16. Nakazawa N. Challenges in the accurate identification of the ideal catheter tip location. J Assoc Vasc Access. 2010;15(4):196-201.
j
Vol 21 No 4
j
JAVA
j
241
R E P O R T
Bedside Chest Radiographs and How Ambiguous Peripherally Inserted Central Catheter Tips Happen: A Case Report Stuart Gordon, MSN, RN, CRNI, VA-BC, RN-BC Legacy Emanuel, Portland, OR
Abstract Verification of peripherally inserted central catheter (PICC) tip location with radiographic imaging is at times challenging. This case report details the experience of 1 patient and the difficulties encountered identifying the PICC tip using bedside chest radiographs. It examines, offers to explain the issues, and reviews the results of 209 chest radiographs post-PICC placement for consistency. Keywords: PICC confirmation radiograph, radiograph gold standard, bedside chest radiograph
Introduction he tip of a peripherally inserted central catheter (PICC) should ideally be in the distal one-third of the superior vena cava (SVC) or cavoatrial junction (CAJ), but correctly identifying placement using bedside chest radiographic imaging is often difficult. The following is a detailed examination of the errors introduced with bedside chest radiographs, followed by a case report. Inaccuracies seen with bedside chest radiographs may lead to unnecessary manipulation or replacement of a PICC; may increase costs to patients; may increase the risk of infection; and may result in the PICC tip terminating, unseen, in a location more distal than the CAJ. The purpose of this article is to bring into focus how bedside chest radiographs fail. The use of bedside radiographic imaging to verify central line tip location and identify placement-related complications is considered by many to be the only acceptable assessment method postinsertion. However, those placing PICC lines have experienced first-hand the difficulties of seeing the tip or accurately judging the appropriate distance to withdraw the PICC when the tip is determined to be in the atria. Members of vascular access teams placing lines are more intimately familiar with the limitations and frustrations than the independent providers who only see the results of a team’s
T
Correspondence concerning this article should be addressed to [email protected] http://dx.doi.org/10.1016/j.java.2016.08.001 Copyright © 2016 ASSOCIATION FOR VASCULAR ACCESS. Published by Elsevier Inc. All rights reserved.
work. With the widespread introduction of electrocardiogram (ECG), ECG/Doppler, and ECG/magnetic PICC tip locators the limitations of radiographs have only been magnified. Accurately identifying the location of a PICC tip with initial placement saves nursing time and patient cost. If, for example, the tip of a PICC is seen projecting below the CAJ institutional policy may require retraction of the line to bring it out of the atria. This requires nursing time to perform the adjustment, it delays use of the line, it requires disruption of the overlying dressing (an intervention that has been shown to increase the risk of a central lineassociated bloodstream infection1), and it will increase the overall cost to patients. Alternatively, if the tip of a PICC is seen at the confluence of the brachiocephalic veins many will remove and replace the PICC, again increasing cost, nursing time, and the risk of infection. This case study suggests initial bedside chest radiograph errors may lead to unnecessary adjustments of PICC lines. Discussion The impediments to accurately identifying the tip of a PICC with a bedside chest radiograph are patient position in relation to the source of radiographic images, whether the radiograph is taken during inspiration or expiration, cardiac motion blur, interreader variability, patient position, and patient body habitus. For the following examples and for the case study it may be helpful to think of radiographic results as shadows. A radiation source will emit high-energy photons that travel through or are blocked by, in this case, the bones, organs, and a central line in the chest on their way to the radiography plate. Parallax error will occur when the radiographic source and the radiography plate are not in 90 alignment. The idea of
2016
j
Vol 21 No 4
j
JAVA
j
237
Central line X-ray source
here in alignment with actual
X-ray plate
Tip seen
locaƟon C
A
Figure 1. Fluoroscopic placement of the port.
image
after
initial
parallax can be visualized by thinking of a right triangle, where the source of the radiograph is point A, the path of photons from the high energy (radiograph) source is side B, and the resulting image of the tip is point C (Figure 1). If the source of photons is not 90 to the plate the path of photons becomes side C and the image of the tip becomes point B instead of the actual tip at point C (Figure 2). Using the equation p2b a ¼ p 2ðpbÞ (solving for side a of a right triangle) if the source of radiographs is 7 either side of 90 (ie, 83 or 97 ) and the distance from point A to point C is 15 cm (average distance from the tip of the catheter to the radiography plate in a 70-75 kg patient) there will be a 2 cm (rounded up) error in identifying where the tip of the catheter appears. Although 2 cm may seem trivial, if the radiograph shows the PICC 2 cm into the atria due to parallax, the result may be an unnecessary retraction. Having the source of radiographs and the radiography plate in correct alignment (as they are set in a radiology department) would all but eliminate parallax; however, taking a patient to a radiology department is not always an option. The ability to see the tip of a PICC is in part dependent on the size of the patient. Adequate penetration of x-rays through tissue is contingent on exposure to the source and larger patients are exposed for longer times. Longer exposure times increase the chance of having cardiac motion and blood turbulence whip the catheter, blurring the image in the same way long camera exposure times will blur pictures. This is often referred to as cardiac motion blur and as the fluoroscopic images in Figures 3 and 4 show, it is not limited to PICC catheters nor is it limited to bedside chest radiographs.
Actual locaƟon
X-ray plate
Central line
Due to parallax Ɵp is seen here
Figure 2. Fluoroscopic image before port revision secondary to palpitations.
238
j
JAVA
j
Vol 21 No 4
j
Figure 3. Fluoroscopic image of a port. Figures 3 and 4 are of the same patientdFigure 3 after initial placement of the port and Figure 4 before port revision secondary to palpitations. Figure 5 shows a PICC, immediately postplacement, whereas Figure 6 is of the same patient 3 days later. This patient’s electronic health record was reviewed to verify that the line had neither been manipulated nor had it migrated, it was cardiac motion that blurred the distal end of the PICC. Several factors contribute to the differences in interpretation of tip location, the first being interreader variability. Different radiologists will use different landmarks when interpreting what they see, and being able to see the PICC tip isdas noted abovedsometimes difficult. There is little consensus as to how well bedside chest radiographic images correlate to the CAJ when using common landmarks. Vesely states, “A catheter tip positioned 3 cm below the right tracheobronchial angle would always be in the SVC,”2 and in a 2008 study Verhey found, “The average distance from the superior vena cavaeright atrial appendage to the cavoatrial junction was 1.8 cm” on chest radiographs in adult patients.3 A similar conclusion was reached by Baskin et al,4 who found “the true cavoatrial junction is located more inferiorly than commonly believed and is not accurately estimated with commonly used imaging landmarks.” The population in that study was predominately pediatric patients lying in a supine position and the authors reviewed computerized tomography (CT) results. Another study using CT found tip placement 4 cm below the carina “will result in placement near the cavoatrial junction.”5 Perhaps the results of these articles is best summed up by Wirsing et al,6 who used transesophageal echocardiography to assess accuracy of bedside chest radiographs and determined, “Reading of a bedside chest x-ray alone is not very accurate to identify intraatrial [central venous catheter] tip location.” They went on to say, “Although bedside [chest radiograph] can be a convenient screening tool for some problems, it cannot serve as a ‘gold standard’ for assessing a [central venous catheter] tip position with respect to the right atrium.”
2016
Figure 6. Three days after placement. Figure 4. Port visualized with fluoroscopy. The port was revised due to palpitations. Cardiac motion blur is present. Three additional factors influencing PICC tip location are arm position, phase of inspiration/expiration, and whether the patient is lying or standing for the radiograph. Respiratory motion was shown to alter central line tip location by 9 mm in 1 study.7 Adduction of the arm from 90 to 0 will move the tip of the catheter 2 cm on average, an effect that is greater
Figure 5. Bedside chest radiograph of a peripherally inserted central catheter, immediately postplacement.
on the right than the left,8 and moving from a supine position to a standing position will cause abdominal contents to shift in a caudal direction an average of 3.2 cm.9,10 Several articles written in recent years have questioned the accuracy of bedside chest radiographs for central line tip placement when compared to ECG systems,11,12 whereas another article13 questioned whether bedside chest radiographs should be considered the gold standard. Challenging old-guard practices have come after gaining a deeper understanding of the limitations of bedside radiographs and those factors reducing accuracy.14 To gain a better understanding of how often the challenges with bedside chest radiographs occur, a research proposal was submitted to our institution’s institutional review board that granted approval for both a case report and a retrospective chart review. Case Report Consider the case of a 49-year-old, 145-kg patient admitted with a chief complaint of sudden onset severe abdominal pain, subsequently diagnosed with gastric perforation. He had surgical intervention, was intubated, and several days into his hospitalization his central line was removed and a PICC was ordered for long-term antibiotics and total parenteral nutrition administration. After the PICC was successfully placed (using external landmark measurements and without the aid of magnetic, ECG, ECG/magnetic, or ECG/Doppler guidance) a radiographic image was taken and read by a radiologist. It was determined that the right PICC catheter tip was in the SVC approximately 6 cm above the CAJ. Both the nurse inserting the PICC and the ordering physician were unable to see the tip of the PICC with clarity and ordered a repeat radiograph. There had been no manipulation of the line in the interval and the radiologist determined that the distal PICC was in the SVC, tip not well seen. The radiologist, now being unable to see the tip, suggested a third radiograph,
2016
j
Vol 21 No 4
j
JAVA
j
239
in which the right arm PICC was visible to the level of the mid-SVC but the location of the tip was ambiguous. A fourth radiograph was ordered, taken, and it was interpreted that the tip of the right arm PICC was near the CAJ. It is worth noting again that there had been no manipulation of the line, and despite 3 prior radiographs, the single radiograph in which the tip was interpreted to be in the ideal location was the radiograph accepted as accurate, an interesting example of confirmation bias, or the tendency to find information that corroborates existing belief. At our institution it is standard for all intubated intensive care unit patients to have daily chest radiographs and for our case report these provide further examples of the challenges with bedside radiographs. Day 1 post-PICC placement it was reported that the central venous catheter was projecting from the right neck with its tip in the plane of the SVC about 4 cm above the CAJ. Day 2 post-PICC placement it was reported that the central venous catheter tip was in the plane of the SVC about 5.3 cm above the CAJ. Day 3 post-PICC it was reported that the right PICC catheter tip was in the SVC at the approximate level of the CAJ. On Day 4 it was reported that the right PICC catheter tip was at the CAJ, and on Day 5 it was reported that the right PICC catheter tip was in the right atrium approximately 2 cm below the CAJ. A quick review of the radiographs showed the tip appeared in different locations, from 6 cm above the CAJ to 2 cm into the atriada total of 8 cm although there had been no manipulation of the line, the patient’s arms were at his side for each radiograph, and the head of the bed was positioned at the same angle. Findings The patient experience noted above is an outlier but illustrates the point that bedside chest radiographs are not infallible. The author reviewed 414 post-PICC placement radiographic images as a routine quality measure. A radiologist interpreted the PICC tip at the CAJ in 110 patients with the tip seen an average of 4.5 cm below the carina, the extremes being 0-10 cm. Only 209 of the 414 patients (50%) had more than 1 radiograph after placement, and these were assessed for consistency in identified tip location, and the electronic health records of each were reviewed to confirm there had been no manipulation or migration of the line that would explain the observed discrepancy. In these serial radiographs the PICC tip was identified in a location that differed from the original in 68 (32%), ranging from 1.5-8 cm. As an example of the extremes in variation, the results of 1 PICC are offered where the tip was visualized “at about the level of the carina, which measures approximately 6.5 cm above the CAJ.” The PICC migrated out 2 cm and to confirm tip location another radiograph was taken after which it was determined that the right PICC catheter tip was in the SVC above the CAJ. Eight days later, with no migration or manipulation of the line, another routine bedside radiograph was taken and read as “the tip is in the SVC approximately 8.5 cm above the CAJ.” Three of the patients above had a CT scan after placement. One of the 3 had 2 radiographs before the CT scan that were
240
j
JAVA
j
Vol 21 No 4
j
reviewed in the interpretation of the second radiograph, which reported that “both images demonstrated the PICC at the level of the carina about 4 cm above the CAJ.” It was noted that the first image was “questionable for the PICC being visible in the right atrium about 1 cm beyond the CAJ.” Two days later the CT scan showed the tip of the central venous catheter was in the lower right atrium. The second patient had the tip of the PICC identified as in the proximal right atrium with radiograph, 4 days later with CT the PICC tip was seen in the proximal SVC. The third patient’s PICC tip was seen on radiograph in the proximal SVC within 1 cm above the CAJ, the next day during the CT scan the tip was seen projecting at the caudal aspect of the right atrium inferior vena cava junction. One patient had a transesophageal echocardiogram (TEE) after PICC placement. The immediate postplacement radiograph showed that the PICC tip was projecting at the CAJ while the PICC was seen extending beyond the tricuspid valve on the TEE. The electronic medical records of the 3 patients with CT scans and the 1 patient with the TEE were reviewed to confirm that there was no manipulation or migration of the PICC. Conclusion As discussed, there is a degree of inaccuracy with bedside chest radiographic imaging and it is hoped that this article opens a discussion about the use of alternative modalities for identifying PICC tip location such as ECG. However, using bedside chest radiographic imaging as a method to gauge accuracy of an ECG device may produce some incongruous results. If alternate methods such as ECG are not an option, having postplacement radiographic images taken in a radiology department when possible will eliminate parallax and may reduce the rate of placements where the tip is not well seen. Disclosures The author has no conflicts of interest to disclose. References 1. Timsit JF, Bouadma L, Ruckly S, et al. Dressing disruption is a major risk factor for catheter-related infections. Crit Care Med. 2012;40(6):1707-1714. 2. Vesely T. Central venous catheter tip position: a continuing controversy. J Vasc Intervent Radiol. 2003;14: 527-534. 3. Verhey P. The Right mediastinal border and central venous anatomy on frontal chest radiography-direct CT correlation. J Assoc Vasc Access. 2008;13(1):32-35. 4. Baskin K, Jimenez R, Cahill A, Jawad A, Towbin R. Cavoatrial junction and central venous anatomy: implications for central venous access tip position. J Vasc Intervent Radiol. 2008;19:359-365. 5. Mahlon M, Yoon H. CT Angiography of the superior vena cava: normative values and implications for central venous catheter position. J Vasc Intervent Radiol. 2007;18: 1106-1110. 6. Wirsing M, Schummer C, Neumann R, Steenbeck J, Schmidt P, Schumer W. Is traditional reading of the
2016
7.
8.
9.
10.
bedside chest radiograph appropriate to detect intraatrial central venous catheter position? Chest. 2008;134(3): 527-533. Pan PP, Engstrom BI, Lungren MP, Seaman DM, Lessne ML, Kim CY. Impact of phase of respiration on central venous catheter tip position. J Vasc Access. 2013;14(4):383-387. Forauer A, Alonzo M. Change in peripherally inserted central catheter tip position with abduction and adduction of the upper extremity. J Vasc Intervent Radiol. 2000;11(10):1315-1318. Nazarian G, Bjarnason H, Dietz C, Bernadas C, Hunter D. Changes in tunneled catheter tip position when a patient is upright. J Vasc Intervent Radiol. 1997;8:437-441. Kowalski C, Kaufman J, Rivitz S, Geller S, Waltman A. Migration of central venous catheters; implications for
11.
12.
13.
14.
2016
initial catheter tip positioning. J Vasc Intervent Radiol. 1997;8:443-447. Gebhard R, Szmuk P, Pivalizza E, Melnikov V, Vogt C, Warters R. The accuracy of electrocardiogram controlled central line placement. Anesthes Analg. 2007;104:65-70. Pittiruti M, Scoppettuolo G, LaGreca A, et al. The EKG method for positioning the tip of PICCs: results from two preliminary studies. J Assoc Vasc Access. 2008;13(4): 112-119. Oliver G, Jones M. ECG or X-ray as the ‘gold standard’ for establishing PICC-tip location? Br J Nurs. 2014;23(19 Suppl):S10-S16. Nakazawa N. Challenges in the accurate identification of the ideal catheter tip location. J Assoc Vasc Access. 2010;15(4):196-201.
j
Vol 21 No 4
j
JAVA
j
241