VOL: 97, ISSUE: 35, PAGE NO: 40
VAL IRVING, RN, RM, BA, is sister/team leader, Neonatal Unit, Liverpool Women’s Hospital
Extravasation of intravenous fluids is said to occur when there is non-intentional leakage of infused fluid into the surrounding tissue, which subsequently leads to tissue damage and skin necrosis. Fluids capable of causing this range from concentrations of glucose greater than 10%, total parenteral fluid which contains potassium and calcium and sodium bicarbonate, through to chemotherapeutic agents such as doxorubicin and vincristine (Gault, 1993).
The reported incidence of these types of injuries vary greatly between population groups and causative agent. Gault (1993) found that extravasation occurred during 0.1-0.7% of all cytotoxic drug administrations, while for patients receiving a course of therapy the risk increased to 4.65%. Infiltration of intravenous fluids - when there is no tissue damage - was found to be 11-23% in paediatric patients and as high as 70% in the neonatal population, although it was thought that tissue damage due to extravasation was considerably lower. Recent figures are hard to find, with the most frequently quoted children’s study being over 20 years old (Brown et al, 1979). This claimed skin loss in 0.24% of peripheral lines sited, but this may have little relevance to the group of neonatal patients cared for today.
Premature infants are now surviving at very low gestational ages and weighing only in the region of 500g, and these neonates are at great risk of sustaining tissue damage as a result of the intravenous therapy required to keep them alive. They have longer exposure time to intravenous therapy until they are able to take all nutritional requirements enterally, their skin is correspondingly immature and easily damaged, there is reduced venous integrity and the median survival time span for intravenous cannulae is only 31 hours (Hecker, 1993).
What is generally accepted is that tissue damage from extravasated fluids does occur and can cause serious injuries that require treatment and may even extend the length of hospital in-patient stay.
Various strategies have been introduced to try and reduce the risk of extravasation injury in preterm neonates:
- Siting of central lines for the administration of total parenteral nutrition and infusions which contain glucose in concentrations greater than 10%;
- Hourly recording of observations at the cannulation site for signs of oedema, firmness or discoloration. This is now routine practice in most neonatal units;
- Securing the cannulae with a transparent dressing to allow unobstructed visibility at the insertion site. This recommendation (Smith, 1985) is more than 15 years old, it is therefore worrying that anecdotal evidence indicates that non-sterile tape or opaque adhesive products are the preferred method of securing the cannulae even today in some units;
- The avoidance of metal cannulae is now standard practice, as the risk of infiltration / extravasation from metal cannulae was found to be much greater due to their inflexibility (Tully et al, 1981);
- The use of pressure sensitive infusion pumps. However, it should be noted that pumps which have preset alarm limits are often not sensitive enough to pick up early pressure changes in these patients. Initially there would be an increase in pressure due to resistance in the vessel, but this falls when the fluids are pumped into the surrounding tissue. As the vessels are very small in the neonate, this change may be very subtle.
Some infusion pumps have adjustable pressure limits that can be altered according to the patient, viscosity of the infused fluid and the length and bore of the tubing used, with a narrow alarm margin of +/-10% advised by the manufacturers. Furthermore, a device with ‘in-line’ pressure monitoring uses the increase in pressure to indicate the degree of resistance - a continual increase would indicate occlusion and therefore require investigation.
However, there are no universally agreed guidelines for pre-term infants on what is an acceptable pressure for any particular weight or gestational age, cannulation site, rate of infusion, viscosity of the infused fluid and so on. This means that decisions regarding alarms tend to be subjective, which renders them virtually ineffective for accurate early detection of infiltration/extravasation of infused fluids.
Differing theories for initial management have been published over the years. Chandavasu et al (1986) suggested that multiple puncture wounds should be made in the affected limb and the extravasated fluid squeezed out, thus reducing the potential for damage.
Beason (1990) recommended cooling the affected area to prevent spread of the irritating fluid and so reduce tissue damage. Conversely, St. Germain et al (1994) felt that heat applied to the area would cause vasodilation and encourage the spread and dilution of the damaging fluid.
The use of steroids were felt to be beneficial by Dorr (1994), but this was disputed by Camp-Sorrell (1998), as there appears to be no inflammatory response to the extravasation of fluids. Flemmer and Chan (1993) recommended the use of glyceryl trinitrate for infants over three weeks old and with intact skin.
Hyaluronidase is an enzyme that temporarily breaks down hyaluronic acid, the ‘cement’ found in the ground matrix of the connective tissue, allowing rapid diffusion of extravasated fluids through the tissues. It has been used since the 1950s for hypodermoclysis and since 1976 for extravasation of intravenous fluids and is said to be free of side-effects.
However, few controlled studies have been carried out in humans. Despite the lack of research, Banta (1992) recommended its use in neonates whenever there was concern about significant tissue damage. Davies et al (1994) presented case histories to demonstrate the use of hyaluronidase and saline flushes in two preterm infants and felt that the outcomes were good, as there was minimal residual scarring. However, this therapy has not gained wide spread support in the UK, and the Royal Pharmacological Society (1999) recommends that it is only used with caution in infants.
There is no universal, national or even regional agreement on the best practice for managing extravasation injuries, and each unit tends to have its own preferred method. This is despite the fact that there is limited evidence to support the use of any particular dressing in the preterm population.
The assessment of wounds in this patient group is as important as in any other group to ensure that the best dressing product is chosen (Miller, 1999). There is no single product that is suitable for all wounds, and the site and size of the injury, the level of exudate, state of healing and possibility of infection must therefore be taken into account.
The sites most commonly used for cannulation in the preterm infant are the dorsum of the hand, using the cephalic or basilic veins, and the dorsum of the foot, using the lateral marginal vein or the dorsal venous arch. These sites are chosen as they are easily viewed, either by eye or by illumination, using a fibre-optic light source. Tissue damage at these sites can cause nerve or tendon damage, which may lead to loss of movement over joints. The superficial temporal vein is also used, but the cannulation site should be above the hair line to reduce the possibility of scarring, should an extravasation injury occur.
Dressings for extravasation injuries must allow a full range of movement to minimise the risk of contractures and at the same time be able to mould to the ankle or wrist areas to create a good seal around the wound.
The area of tissue damage may only cover as little as 1cmx1cm, but in a very preterm infant this could involve the entire dorsum of the hand and so be of great significance. Dressings must be small enough, or be able to be cut to size to fit well, without reducing the potential to absorb exudate.
This is produced due to the increased permeability of the capillary vessel walls in response to tissue damage and consists of proteins, nutrients and growth factors.
Exudate acts as a medium for phagocytic activity and is most prolific during the inflammatory phase of wound-healing. It is just as difficult to determine the level of exudate in these tiny wounds as with others in the adult population, and so subjective opinions of low/medium/large are usually recorded.
- Necrotic: the surface of the wound may initially be covered with devitalised tissue which is still unbroken if the area of damage is away from the cannulation site. As with other wounds, this tissue has to be removed before healing can begin (Tong, 1999) (Fig 1);
- Slough: this is composed of dead white cells (neutrophils and macrophages) and can be mistaken for pus by the inexperienced carer who has failed to notice the lack of other signs of infection, such as redness, swelling, heat and loss of movement (Fig 2);
- Granulating: granulation tissue tends to develop relatively quickly once the wound bed is clean and gives the wound a red appearance. This is a very vascular tissue and bleeds easily if damaged. Great care should be taken during the dressing change or handling of the neonate to prevent this happening (Fig 3);
- Epithelialisation: this is thought of as being the last stage of healing when the epithelial cells move across the wound from the edges towards the centre. The wound bed now has a pink appearance (Miller, 1999) (Figs 4 and 5).
Signs of infection
As previously highlighted, these are no different to any other population group, but the neonate cannot tell of the pain or discomfort. Pre-term infants react to infection by having increased apnoeas and bradycardias with associated saturation drops, an increased oxygen requirement, becoming intolerant of enteral feeds and/or temperature instability.
A number of modern wound contact materials have been found to be of use in the management of extravasation injuries in preterm neonates.
The use of a thin hydrocolloid was demonstrated by Young et al (1996) to be beneficial in the treatment of extravasation injuries in the pre-term population (Fig 6). This type of product has been used for some time as a skin barrier to prevent epidermal stripping due to tape and adhesive products and was subsequently found to possess many of the features of an optimum dressing for this patient group (Table 1). It also has the added benefit of an extended wear time of seven days which is achievable during the granulation and epithelialisation stages of wound-healing.
However, as the dressing is frequently very small its absorption properties are limited and during the inflammatory stage of wound healing, when the level of exudate is greatest, daily dressing changes may be necessary to prevent ‘strike-through’ (Fig 7). Anecdotal evidence suggests that this frequency of dressing change does not cause damage to the surrounding skin, as the adherence of the dressing is gradually reduced by the pressure of the exudate, and tension is not required to remove it. The use of a hydrosorbitive dressing in combination with a hydrocolloid increases the wear time during this phase, but more work is needed to establish its full benefit in the neonatal population.
Thomas et al (1987) found that the use of a hydrogel created a moist environment and led to reduced scarring when used on extravasation injuries in preterm infants. Due to the high ambient temperatures these neonates require to maintain their core temperature (in incubators up to 40°C or room temperatures between 24-26°C) the gel has to be placed in a plastic bag to prevent it drying out. There are no specific bags manufactured for this purpose and products such as urine collecting bags have to be adapted, which has the potential to reduce their sterility. The gel is either applied directly to the injury or added to the bag after the limb is inserted and then sealed with tape or a film dressing. The removal of the latter involves a risk of epidermal stripping to the already damaged limb, so great care must be taken (Fig 8).
When applying the hydrogel using a bag neonates can move their fingers or toes, but the overall weight often prevents them from moving their entire limb and the dressing may require splinting to support it. The gel can absorb a moderate to large amount of exudate but once the necrotic tissue begins to separate by the process of autolysis the gel will become discoloured and this may be upsetting for the parents to view. The bag can remain in situ for two to three days if there is no sign of wound infection, when it would then require a daily change.
Hydrogels can be of benefit for dorsal limb injuries where involvement of the digits means that a good seal cannot be obtained or where the exudate is so large that repeated ‘strike-through’ necessitates frequent changes of a hydrocolloid dressing, for example. The excessive handling this would entail could cause a deterioration in the overall condition of the neonate, with a resulting increase in oxygen or ventilation requirements, and should be avoided if possible.
Although there is no published evidence for the use of film dressings as a wound contact material in the preterm population, in practice these products can be found in many neonatal units. The benefits of their use are that the wound bed can be seen at all times and that epidermal water loss is reduced, compared to leaving the wound exposed.
Harpin and Rutter (1983) showed that the transepidermal water loss through the intact skin of an infant of 26 weeks’ gestation was up to 1560g/m2 per 24 hours, while at term it fell to only 240g/m2 per 24 hours. Studies in the adult population (Thomas, 1990) have shown that the moisture vapour transpiration rate through uncovered wounds is more than 10 times as great as when the skin is intact, at 3,400-5,200g/m2 per 24 hours, and although no work has been carried out on preterm neonatal wounds this rate of increase is likely to be mirrored and could lead to fluid imbalance and temperature instability if not addressed.
The moisture vapour permeability of film dressings in common use range between 846 g/m2 per 24 hours and 862 g/m2 per 24 hours. This means that exudate from the wound may be retained under the film dressing. While a moist wound environment is beneficial, excessive exudate can lead to maceration of the surrounding skin and, although not generally advocated, aspiration of this fluid may be required.
With the very preterm neonate there is the increased risk of trauma to the surrounding skin if adhesive dressings or products are not removed by the horizontal stretch method. Many staff are not fully aware of the damage that can be caused if the dressing is removed incorrectly.
The injury caused by extravasation of intravenous fluids is an iatrogenic injury and as such needs to be formally recorded as an incident. If there was concern about the patency of the cannula, or any swelling around the site before the extravasation, there needs to be a detailed account of the action taken at the time. If the unit policy dictates that hyaluronidase or other drug therapies are used then this must be carried out as soon as possible after the injury is identified.
There needs to be clear documentation of the causative agent, the site and size of the injury, an assessment of the wound and the rationale behind the choice of dressing product applied, and a care plan must be started for the treatment. A photograph of the wound with some form of scale included should be attached to the case notes, and the parents should be informed at the first opportunity.
With each dressing change the state of the wound bed should be described and changes in the size of the wound recorded, along with details of the replacement dressing and planned review date.
If the wound has not completely healed by the time the infant is ready for discharge, the parents and the community staff can be involved with the dressing changes so that continuity can be maintained.
If the infant is being transferred from a regional unit when intensive care is no longer required, there needs to be good liaison between both sets of staff so that the planned care is completed with optimum healing and minimal residual scarring. Copies of the wound care guidelines, wound assessment and care plan documentation should also be sent with the discharge information.
With increasing litigation it is important that detailed accurate documentation is kept at all stages of the wound-healing process and, although this does not detract from the injury which has occurred, it will show that the treatment provided was based on appropriate experience and knowledge.
Involvement of plastic surgeons
In cases of extravasation injury in term infants or older children it is best to involve the plastic surgeons at an early stage. However, for the very preterm neonate this is not a beneficial option. Only in extreme cases is immediate repair surgery necessary and the size and immaturity of these infants can pose problems that plastic surgeons may not be prepared for, and suggested treatments are often not appropriate for this patient group.
Unfortunately skin and wound care in preterm neonates has not progressed to the same extent as other areas of medically led neonatal practice. This is therefore an area where nurses can develop a role similar to that in adult care. There does not appear to be any nurse-led research into preterm tissue viability issues being conducted in the UK. The American literature suggests that there is greater interest in this area in the USA (Maguire, 1999; Munson et al, 1999). However, this is not reflected in practice, and care here is rarely research-based and often non-standardised.
There is a lack of information on the incidence and prevalence of extravasation injuries, research-based treatments and wound products suitable for neonatal patients. The rate of survival of very preterm neonates is constantly improving, and this aspect of their care must also improve. These tiny individuals require specially designed effective pressure-sensitive equipment for the early detection of extravasation of intravenous fluids, effective treatments and optimum level of wound care to ensure minimal residual scarring.
Liaison between specialists in the field of tissue viability should be encouraged and their input included in all postregistration neonatal courses so that preterm infants receive the same, if not better, level of care in this area as other patient groups. The fact that these patients may be relatively few compared to other groups does not mean they should be ignored: to provide intensive care can cost at least £1,200 per day, so effective wound care must be part of that service.