Send letter to journal:
Re: Pericardial effusions caused by neonatal long lines - how about management ?

Dear Editor

I congratulate Beardsall et al. on their impressive, albeit retrospective collection of data.[1] Their important conclusion that neither extra-atrial position of the catheter tip, nor imaging with contrast, are reliable measures to prevent pericardial effusion (PCE) means that we will have to live with this threat. This inevitably redirects attention from the aspect of prevention to the question of how to guarantee optimal management, which may be the only way to reduce mortality from this complication. As Beardsall et al. [1] focus on epidemiology and prevention, an obvious wealth of data on the issue of management has been hidden rather than fully exploited in the text and Figure 1.

As stated by Menon 2, PCE is nearly always bloodless, probably representing a pericardial accumulation of infusate. This gives the opportunity not only to diagnose the condition simply by aspirating typical infusate (with high glucose concentration, and turbidity in case of lipid admixture) from the long line, obtaining a volume which by far exceeds the dead space of the catheter; but it also allows us to reduce the volume of the effusion without having to perform a pericardial tap with its own risks. Since nearly two thirds of affected neonates present with sudden cardiovascular collapse,[2] the time required to obtain an Xray and / or echocardiography to diagnose the condition may reduce the chance of survival. The instinctive reaction to remove immediately the long line responsible for the patient’s critical condition may not be advisable in this situation. A preferable option would be to use the catheter in its actual position to aspirate as much of the effusion as possible, and thereafter to withdraw it until blood can be aspirated with ease, indicating an intracardiac / intravascular (re-)position of its tip, providing the option to use this line for central venous administration of inotropes if required. This procedure can be accomplished rapidly, does not interfere with resuscitation, and may reduce the number of cases who ultimately need a pericardial tap. Interestingly, Beardsall et al. reported that all cases treated “conservatively” survived.[1]

Clearly, the data collected by Beardsall et al.[1] deserve to be analysed in more detail in order to evaluate the success of different management strategies. The resulting message could be more encouraging than the preliminary conclusion that little else can be done to prevent this life-threatening complication.

References

(1) Beardsall K, White DK, Pinto EM, Kelsall AWR. Pericardial effusion and cardiac tamponade as complications of neonatal long lines: are they really a problem? Arch Dis Child Fetal Neonatal Ed 2003;88:F292-295.

(2) Menon G. Neonatal long lines. Arch Dis child Fetal Neonatal Ed 2003;88:F260-262.

Send letter to journal:
Re: Long line tip position and cardiac tamponade

Dear Editor

We read with interest the article on neonatal long lines[1] and the accompanying editorial[2] published in Archives recently. One of the important conclusions reached was that 76% of cases of pericardial effusion/cardiac tamponade occurred in units who aimed to position the tip in the vena cavae. So simply following the Department of Health guidelines to avoid the cardiac chambers may not be sufficient to avoid this complication, although accurate positioning of long lines clearly remains important.

We undertook an audit to look at the position of the long lines inserted in our unit (Hammersmith Hospital, London) over a 7 months period from January to July 2002. All the 66 long lines in 61 neonates, (Epicutaneo-Cava-Katheter - Vygon, UK) placed during the study period were reviewed on the Picture Archiving and Communication System (PACS) using the magic glass (sharpen and invert) tools as surrogate contrasts. All the films with a long line were looked at to identify malposition and/or migration. 41 of the 66 long lines were outside the heart in the initial Xray and remained outside henceforth. 15 lines were inside the heart on the initial Xray but were re-positioned and remained outside thereafter. 10 lines were inside the heart for varying periods. 7 of these 10 lines were inside the heart from the time of insertion for between 2 to 10 days before being withdrawn or removed. The remaining 3 lines were in an acceptable position on the initial Xray but subsequently migrated remaining inside the heart for 7, 7 and 14 days respectively before being withdrawn or removed. No instances of pericardial effusion or cardiac tamponade were observed during the audit period.

Despite attempting to place all long lines outside the heart 10/66 (15%) were inside the heart for times ranging from 2 to 14 days. Long line position should be carefully checked on insertion and reviewed regularly as a small but significant number of lines migrate prior to removal.

References

(1) G. Menon. Neonatal Long lines. Arch Dis Child Fetal Neonatal Ed 2003;88:F260–F262.

(2) Beardsall K, White DK, Pinto EM, et al. Pericardial effusion and cardiac tamponade as complications of neonatal long lines: are they really a problem? Arch Dis Child Fetal Neonatal Ed 2003;88:292–5.

Send letter to journal:
Re: Neonatal long line and cardiac tamponade: a rare but important complication

Dear Editor

The original article from Cambridge by Beardsall et al.[1] reminds us of the problems of neonatal long lines and also presented the outcome of managing cardiac tamponade in 82 cases.

In my unit in a busy district general hospital with 3 NICU cots with an average of 3300 deliveries per year and with 10% of this admitted to the neonatal unit, we have had a share of this rare problem. In the last 10 years we have regularly inserted neonatal catheters for parenteral feeding and also inserted umbilical catheters for fluid resuscitation when venous access is difficult. On average, 20-50 catheter insertions occur every year. Over this period we have experienced the unfortunate death of a term newborn who had a portex catheter inserted for the infusion of hypertonic dextrose solution for managing persisting hypoglycaemia. An unexpected collapse did occur which made us suspect cardiac tamponade. Active resuscitation was carried out but was regrettably unsuccessful.

The DOH inquiry[2] into the cases in greater Manchester has highlighted the need for vigilance and the ensuing debates has created more questions which calls for appropriate research to establish the right answers. Although this is a rare complication, (1 case in 10 years from my unit), it can result in fatality. Cardiac tamponade remains a serious complication that must be considered always and should leave no room for complacency.

We have since strengthened our good practice guideline for the use of neonatal catheters especially after the DOH reviews. We run through the following points each time a central line insertion is contemplated:
(a) Decide if catheter use was the best mode for venous access for the neonate.
(b) The person inserting must be experienced or adequately supported by an accomplished colleague.
(c) When a decision to insert a catheter is taken an informed consent is obtained from parents. The need for the insertion is discussed and complications detailed.
(d) After placement, contrast is used to locate position and this is adequately documented.
(e) Neonates with indwelling catheter have detailed review of line care and splinting of limb is encouraged to minimise migration. Any sudden deterioration in clinical state is flagged as possible cardiac tamponade (vigilance).
(f) The length of stay for catheters is constantly reviewed and removed as soon as alternative means for nutrition is established.
(g) Consultant must be involved in the decision to insert neonatal catheter.

There is need to determine the best position for indwelling catheters, the type of catheter material, and best way to prevent migrations. The DOH report has highlighted this issue and made suggestions based on limited evidence. In this review, Beardsall et al showed that 50 out of 60 patients survived after management of this complication. Therefore vigilance and prompt action to treat is essential and units performing this procedure must be able to provide emergency treatment for cardiac tamponade.

References

(1) Beardsall K, White D, Pinto EM et al. Pericardial Effusion and cardiac Tamponade as complications of neonatal long lines; Are they really a problem? Arch Dis Child Fetal Neonatal Ed. 2003;88;F292-295.

(2) Department of health Review of deaths of 4 babies due to cardiac Tamponade associated with the presence of central venous catheter. London:HSMO, 2001.

Send letter to journal:
Re: Re: Pericardial effusion and cardiac tamponade as complications of neonatal long lines

Dear Editor

I wish to make a comment about the interpretation many readers could have taken from the paper of Beardsall et al. which mentioned that there was no difference in the cardiac tymponade in babies whether contrast was used or not. In the survey only 26% of units routinely used IV contrast and 43% of cases occurred in units where contrast had been used to confirm the tip position although they did not clarify in this group of 43% whether they used silastic or polyurethane type of long line.

Although the survey relates only to incidence of cardiac tymponade, other complications of long line like extravasation in tissues and pleural effusion are not uncommon. The publication of the case report by Makawana et al1 very nicely demonstrated the usefulness of contrast in identifying the tip of silastic long line as the long line in their case report was obviously not visible beyond SVC-right atrial junction without contrast and was actually found to be entering the retroperitonial collection on use of contrast after the complication had occurred. I am sure there will be several unpublished cases in every unit with similar experiences of extravasations in different tissues.

I would also like to make a similar mention about their statement in the paper about the type of long line used. The authors divided the long lines in to two groups, viz; Medex and Vygon and stated that 35% of cardiac tymponade was associated with Medex lines as compared to 54% with Vygon lines. This may again be a bit misleading for some of the readers as the Vygon company markets both types of long lines, i.e. silastic and polyurethane. The polyurethane ones are similar to the ones made by Medex. Although no clinical significance was deduced from the data, this is quite a significant observation as the polyurethane lines are completely opaque on X-rays and so do not need contrast for their visualisation while the silastic ones are not completely opaque and their visualisation on X-rays is dependent on several factors like exposure of the film, edema of the subcutaneous tissues, opacification of the lungs due to any reasons, e.g. consolidation, pleural effusion etc2.

Therefore, the use of contrast for silastic long lines is quite important and this fact was also emphasised in the recent paper by Odd et al3.

References

1. Makwana N., Lander A., Buick R. and Kumararatne B. Unusual complication of venous line in a neonate. Arch Dis Child Fetal Neonatal Ed 2003; 88: F440

2. Masand M. Long Lines in neonates and complications [electonic response to Makwana et al. Unusual complication of a central venous line in a neonate] archdischild.com 2003 http://adc.bmjjournals.com/cgi/eletters/fetalneonatal;88/5/F440#245