LESLIE SCHAAF TREAS, RNC, MSN, C N P B A R B A R A L A T I N I S - B R I D G E S , R N , MS, P H E D
Emacy of Hepa& in Prnf@mal Vmous Infusion in Neonates Objective: To determine the eficacy of continuous, low dose heparin infusion in prolonging peripheral venous catheter patency in neonates. Design: Randomized, prospective study. Setting: Level III neonatal intensive-care unit. Participants: 113 neonates requiring IV therapy. Interventions: The neonates were randomly assigned to heparin (n = 63). The heparin group received '/z unit of heparin per milliliter of continuous intravenous infusate or intermittent heparin flush. Main outcome measures: Low-dose heparin infusion does not make a difference in the duration of peripheral venous catheter patency. The incidence of catheter-related complications in the heparin group is the same asfor those receiving no heparin. Results: The mean duration of catheter patency was 62.75 hours in the heparin group and 27.3 hours in the no-heparin group ( p = .0001). The occurrence of infection, bleeding, and extravastation injury was zero in the sample studied. The incidence of phlebitis was 18 cases in 132 (13%) in the heparfn group and 13 cases in 122 (10%) in the no-heparin group. According to chi-square analysis, the difference between groups was not statistically signgcant ( p > .05). Conclusion: Lowdose heparin infusion in peripheral venous catheters in neonates increased the duration of intravenous catheter patency ( p = .0001) without increased risks of bleeding, infection, phlebitis, or extravasation injury.
ecause nurses spend a significant amount of time inserting intravenous catheters for their patients, less frequent intravenous starts would allow nurses time to meet other needs of patients and their families. A more effective method of maintaining intravenous catheters in neonates is needed to reduce the identified stress factors associated with the insertion of intravenous catheters. Prolonging intravenous patency without increasing other risks to the newborn is beneficial to the patient and the caregiver.
Purpose Intravenous catheter insertion is time consuming and associated with significant risks to the neonate. Heparin has been used widely in adult patients to prolong the patency of the intravenous catheter, but few data exist regarding the use of low-dose heparin to maintain peripheral intravenous catheters in the neonate.
Intravenous Catheter insertion is time consuming and associated with significant risks to the neonate.
The purpose of this study was (a) to determine whether a low-dose heparin solution prolongs patency of the peripheral venous catheter in neonates and (b) to compare and contrast the nature and incidence of complications of infusion (e.g., phlebitis, infection, extravasation, bleeding) in neonates who receive low-dose heparin in their continuous intravenous infusates and those who do not.
Related Literature
Accepted: September 1991
Continuous low-dose heparin has been used widely to maintain the patency of central and peripheral intravenous catheters on the assumption that the low concentration of heparin accomplishes this goal without causing hemorrhagic complications (Rajani, Goetman, Wenneberg, Turner, & Abilgaard, 1979). Controlled studies in adults show the efficacy of heparin in reducing thrombophlebitis in peripheral (Gillhooly, Lindenberg, & Reynolds, 1986; Tanner, Delaney, & Hennessy, 1980) and central vein infusion sites (Brismar, Hardstedt, &Jacobson, 1982; Daniell, 1973; Fabri, Mirtallo, & Ruberg, 1982; Hoar, Wilson, Mangano, & Avery, 1981) and in prolonging the time until infiltration (Gillhooly et al., 1986).
214 J O G N N
Volume 21 Number 3
Heparin in Peripheral Venous Infusion
Heparin added to intravenous infusates more than doubled the duration of peripheral catheter patency without increasing risks to the neonate.
Studies in newborns show the benefits of both intermittent and continuous heparin infusion in preventing clot formation, reducing the incidence of line-related complications (Horgan et al., 1970), and prolonging the patency of umbilical vessel catheters without increasing the incidence of clinical bleeding (Bosque &Weaver, 1987; David, Merten, Anderson, & Gross, 1981; Gillhooly et al., 1986; Kitterman, Phibbs, & Tooley, 1981; Rajani et al., 1979). Few data exist, however, regarding the use of heparin in maintaining peripheral venous catheters in neonates. Alpan et al. (1984) conducted a controlled study of heparinization of total parenteral alimentation solutions administered through peripheral veins in premature infants. Heparin was found to double the duration of patency of intravenous catheters and to reduce significantly the incidence of phlebitis. The sample group, however, was small and incompletely controlled for several confounding variables relating to catheter life, the incidence of phlebitis, or both. N o data were found in the literature reviewing the use of heparin in peripheral dextrose and electrolyte solutions. Since this kind of fluid is commonly used in the sick neonate’s first days of life, the need to study this aspect of intravenous therapy was compelling.
Methods Subjects The convenience sample consisted of neonates in the level I11 neonatal intensive-care unit of a major teaching hospital. All of the neonates required peripheral intravenous therapy for routine medical care, regardless of birth weight or gestational age. Neonates were assigned to an experimental or control group based on the sequential number routinely assigned by the admitting office of the hospital. The control group included 49 neonates with 1 2 2 peripheral intravenous catheters to which no heparin was added. Saline was added to these intravenous catheters as an irrigating solution for intravenous insertion and intravenous locks. The experimental group included 63 neonates with 132 intravenous catheters to which one-half unit of heparin per milliliter of intravenous infusate and irrigating solution was added. All of the neonates in the study received intrave-
May/June 1992
nous therapy for varying periods and required a series of peripheral venous cannulations. Only the first three intravenous catheters per neonate were included in the study. Subjects excluded from the sample were those with (a) central line infusion (umbilical vessel or percutaneous central line), because heparin is routinely added to fluids administered through these catheters; (b) clinically suspected or documented bleeding diatheses; or (c) peripheral dopamine infusion. If a vasopressor was infused during the study, that intravenous catheter was eliminated from the data collection. ~~~~
Exclusion criteria included central line infusion, bleeding diathesis, and vasopressor use.
Procedure Registered nurses in the neonatal intensive-care unit who were assigned to the care of study subjects inserted the intravenous catheters. The nurses were randomly assigned to patients in both study groups. The intravenous insertion technique was consistent with unit protocol. The most suitable site available was permitted. The intravenous catheters were 24-gauge, % inch Insyte-W (Deseret Medical, Inc., Beckton & Dickinson Co., Sandy, UT). A clear, nonocclusive dressing (Op Site, Johnson and Johnson, New Brunswick, NJ, or Tegaderm, Medical Surgical Division, Saint Paul, MN) was used at the site to stabilize the intravenous catheter. N o antibiotic ointment or other dressing was used. The intravenous solution, in-line filter, and intravenous tubing were changed every 24 hours; medications were uniformly diluted, and blood products and medications were infused using a syringe infusion pump, according to unit protocol. The insertion site was inspected for complications at least every 2 hours by the registered nurse assigned to the neonate’s care. The intravenous line was removed if signs of local infection, phlebitis, or extravasation developed. The bedside or staff nurse recorded a description of the site when the line was removed. All nurses in the unit were instructed in how to define the characteristics of each intravenous complication before and intermittently during data collection. Moreover, a list of definitions of terms was provided to each nurse to ensure standard classification of intravenous site condition at the time of removal of the line. The definition of terms was located on each subject’s bedside clipboard and in various locations in the unit. Each nurse was trained to watch for signs of intravenous-related infection, extravasation, bleeding, and
J O G N N
215
C L I N I C A L
S T U D I E S
phlebitis. If signs of local infection developed, the cannula was cultured; if signs of local or systemic infection developed, two peripheral blood cultures were drawn from separate sites. The nurses were to observe for clinical signs of bleeding, i.e., prolonged bleeding from puncture sites and bloody endotracheal secretions. To detect intraventricular hemorrhage, routine cranial ultrasound was performed on all neonates born at or before 34 weeks’ gestation. All results were interpreted by the same radiologist, who had no knowledge of which group any given subject was assigned to.
Results The heparin and no-heparin groups shared several characteristics. The admission diagnoses of patients in both groups commonly included prematurity, multiple gestation, respiratory distress, suspected or documented sepsis, apnea, hyperbilirubinemia, anemia, and feeding intolerance. The birth weights and gestational ages of patients in the heparin and no-heparin groups were clinically comparable (see Table 1). The heparin additive was the only covariate that was statistically significant in influencing catheter patency.
The location of the intravenous catheter was comparable between the two groups. The hand or arm was the most common site for both groups. Other areas for catheter placement were similar between the groups (see Table 2). Infiltration was the most common reason for removing the intravenous catheter for both groups. Phlebitis was the second most common reason for discontinuing the catheters in the neonates (see Table 3).
Heparin (N
=
63)
Birth weight range Birth weight mean
8954,984 g 2,300 g
1,015-3,495 g
Gestational age range Gestational age mean
(SD, 806 g) 29-43 weeks 34.7 weeks (SD, 3.2 weeks)
(SD, 595 g) 28-42 weeks 33.8 weeks
216 J O G N N
Scalp Antecubital or saphenous Hand or arm Foot or leg
1,977 g
(SD, 3.0 weeks)
No heparin (N = 122)
Heparin (N = 132)
12 (9.84%) 12 (9.84%)
16 (12.12%)
73 (59.84%) 25 (20.49%)
24 (18.18%) 67 (50.76%) 25 (18.94%)
The no-heparin group received more medication than the heparin group (see Table 4). This difference is statistically significant ( p < .05). The heparin group received more total parenteral nutrition, more lipids, more calcium, and higher concentrations of glucose. Both groups received an equivalent number of blood transfusions (see Table 5). One reason for this study was to compare and contrast the duration of catheter patency between two groups of patients, one of which received low-dose heparin in the continuous intravenous fluid and the other of which received no added heparin in the infusate. The mean duration of patency for the 132 intravenous catheters in the heparin group was 62.75 hours; the mean duration of the 122 intravenous catheters in the no heparin group was 27.3 hours. Using the t test, this difference in duration of catheter patency was statistically significant ( p = .0001; t = 7.56). To determine whether the covariates (i.e., lipids, total parenteral nutrition, calcium, blood transfusion, glucose concentration, medications infused through the line) affected the duration of patency, the Cox proportional hazards model was calculated. This nonstepwise reTable 3. Comparison of Reason f o r Catheter Removal Between Groups Reason
Table 1. Birth Weight and Gestational Age by Group No heparin (N = 49)
Table 2. Comparison of Intravenous Sites Used Between Groups
Accidental dislodgement Elective discontinuation Infection Infiltration Leaking Mechanical Phlebitis Plugged
Volume 21 Number 3
No heparin (N = 122)
Heparin (N = 132)
5 (4.10%)
7 (5.30%)
8 (6.56%)
18 (13.64%)
0 (0)
86 (70.49%) 5 (4.10%) 2 (1.64%) 13 (10.66%) 3 (2.46%)
0 (0)
77 9 1 18 2
(58.33%) (6.82%) (0.76%) (13.64%) (1.52%)
Heparin in Peripheral Venous Infusion
Table 4. Comparison of Medications Infused Between Groups
Medicatton None Ampicillin and gentamicin Caffeine Ampicillin, gentamicin, and caffeine Ampicillin and caffeine Ampicillin Gentamicin Other
No hepartn (N = 122)
Heparin (N = 132)
56 (45.90%) 51 (41.80%)
88 (66.67%) 28 (21.21%)
3 (2.46%)
3 (2.27%)
3 (2.46%) 3 (2.46%) 5 (4.10%)
4 (3.03%)
0 (0) 1 (0.82%)
1 (0.76%)
1 (0.76%)
5 (3.79%) 2 (1.52%)
gression statistic was calculated to determine whether the covariates that were measured simultaneously affected the duration of catheter patency significantly. The heparin additive was the only covariate that influenced catheter patency significantly ( p < .OOOl). The study also examined catheter-related risks associated with the use of heparin. Chi-square analysis revealed no significant difference in the frequency of phlebitis in the first three intravenous catheters in either the heparin or the no-heparin group. The frequency of phlebitis in the no-heparin group was lo%, compared with 13%in the heparin group. The difference was not significant ( p > .05), according to Yate's correction for samples with 1 degree of freedom. Heparin infusion in the study was 0.8-3.2 units/ kg/hr. This low-dose infusion of heparin did not produce clinical symptoms of bleeding in the heparin group, and no incidence of bleeding was reported in the no-heparin group. The frequency of catheter-related infection in each group was zero, and the frequency of extravasation injury in each group was also zero.
tion of patency. The difference between the groups is significant ( p = .05). The groups received equivalent amounts of blood. Further research is needed to determine the effects of medication and infusates. The subjects had no infections during the study. Infection related to the catheter has complicated up to 8%of venous cannulations (Fuchs, 1971; Maki, Goldmann, & Rhame, 1973) and is said to result from many factors, including prolonged catheterization (Banks, Cawdreg, Yates, Harris, & Kidner, 1970; Birnbaum, 1981; Collins, Braun, Zinner, & Kass, 1968; Fuchs, 1971; Lowenbraun, Young, Kenton, & Serpick, 1970). Catheters that remained in situ for more than 72 hours in this study, however, caused no increased incidence of infection in neonates. In a study examining the life span of 5,000 intravenous catheters in neonates, the type of medication infused (except pancuronium bromide) did not have a discernible effect on the functional duration of patency (Johnson & Donn, 1988). None of the patients in this clinical study received pancuronium bromide. Phlebitis is known to complicate up to 18%of peripheral venous catheters (Batton, Maisels, & Applebaum, 1982; Tulley, Friedland, Baldini, & Goldman, 1981), but the frequency of phlebitis in this study sample was only 13% in the heparin group and 10%in the no-heparin group. The frequency of extravasation injury was zero. These findings of low frequencies of infection, extravasation, and phlebitis may be related to the Hawthorne effect. Since the nurses who inserted the intravenous catheters and assessed the intravenous site knew they were participating in an experimental study, they may have been more conscientious in their aseptic technique and assessment of intravenous sites to comply with the well-defined study protocol. Table 5. Variables That May Influence Duration of Catheter Patency
Discussion Vartable
The infusion of calcium, lipids, total parenteral nutrition, blood, medications, and high-concentration glucose solutions may shorten the catheter life of a peripheral intravenous line. In this study, the heparin group received more calcium, total parenteral nutrition, and lipids, as well as higher glucose loads, than did the no-heparin group. Although the heparin group received greater amounts of substances that may affect vascular integrity, the duration of patency more than doubled (see Table 4 ) . Conversely, the no-heparin group received more medications through the catheter, which may have been related to a shorter dura-
May/June 1992
No heparin (N = 49)
Total parenteral nutrition 1 (2.04%) Lipids 1 (2.04%) Glucose (high concentration) 2 (4.08%) Calcium 47 (95.92%) Blood 1 (2.04%) Medications 28 (57.14%)
Hepartn ( N = 63)
P
13 (20.63%) 0.003 8 (12.70%) 0.040 1 1 (17.46%) 0.028
63 (100%)
0.106 (ns)
0 (0) 0.255 (ns) 24 (38.10%) 0.045
Chi-square measurement with Yate's correction ns: not significant.
J O C N N
217
C L I N I C A L
S T U D I E S
The researcher was unable to control the activity level of the neonates, which could have an effect on the duration of catheter patency. The exclusion criterion of an umbilical catheter inherently limited how acute the study group’s medical condition would be. The researcher was unable to standardize the site used to one location. Staff nurses chose the most appropriate site for each neonate, and coincidentally, the intravenous sites of the groups were found to be similar.
Nursing Implications
Less frequent intravenous starts are benejicial in minimizing exogenous sources of stress that may lead to physiologic compromise of the sick neonate.
As the patient’s advocate, the nurse may promote the former method of intravenous therapy because less frequent intravenous starts may decrease neonatal and parental stress and lessen the potential for infection and other associated deleterious physiologic consequences of intravenous reinsertion. In addition, less frequent intravenous insertions may represent a cost savings to the family and the health-care facility because fewer materials for insertion and less nursing time would be needed.
The insertion of an intravenous catheter in a newborn often can be difficult because of a reduced number of available veins to cannulate. The procedure can be technically challenging to perform on small vessels, especially on neonates who are sick and have friable veins or poor perfusion. The insertion can cause pain, which can lead to deterioration in the neonate’s condition. Furthermore, the pain response associated with intravenous cannula reinsertion may be associated with a transient rise in the neonate’s blood pressure, a known risk factor for intraventricular hemorrhage (Volpe, 1987). These sick neonates have a diminished capacity to adapt to exogenous stressors in the environment, including noise, temperature fluctuations with handling, and procedures. A sick neonate’s response to overstimulation is manifested by apnea and bradycardia, cyanosis, hypertension and hypotension, hypoxemia, hyperglycemia, elevation of intracranial pressure, and feeding difficulties (American Academy of Pediatrics, 1974; Coolinge & Aranda, 1984; Gottfried & Gaiter, 1985; Long, Phillip, & Lucey, 1980; Norris, Campbell, & Brenbert, 1982; Spiedel, 1982). Complications of intravenous catheter insertion in neonates include (a) hypothermia due to exposure to ambient room temperatures during the insertion of the intravenous catheter, (b) hypoglycemia related to the interruption of glucose infusion, (c) sleep deprivation and hypoxia from the stress of insertion, and (d) increased infection potential with repeated venipuncture (Long et al., 1980; Norris et al., 1982; Spiedel, 1982). In caring for the compromised neonate, the neonatal nurse attempts to minimize exogenous sources of stress and prevent further physiologic compromise. Less frequent restarts of the intravenous infusion are beneficial. Continuous intravenous infusion with heparin added in the peripheral venous catheter prolonged the patency of the intravenous line an average of 62.7 hours. Solutions with no heparin added in the infusate remained patent for an average of 27.3 hours.
Alpan, G., Eyal, F., Springer, G., Glick, B., Groder, K., & Armon, L. (1984). Heparinization of alimentation solutions: A controlled study. Pediatrics, 74, 375-378.
218 J O G ”
Volume 21 Number 3
Recommendations f o r Future Research Replication of this scientific study is needed to validate or dispute its findings. The use of heparin solution versus saline flush for the irrigation of intravenous locks in neonates may be compared. Conflicting data are reported in the literature about adult subjects and the efficacy of heparin irrigation solution in maintaining the patency of intravenous locks. The study could be replicated by comparing the patency of the intravenous catheter without heparin in the infusate to the patency of catheters with the infusion of 1 unit per milliliter of solution. The dosage used in this study examined the effect of ‘/z unit per milliliter of infusate. In many neonatal intensive-care units nationwide, however, the addition of 1 unit of heparin per milliliter is standard. The effect of an increased dose of heparin may influence the duration of patency. The study could be replicated by using a pediatric sample. Few data exist regarding the use of continuous heparin infusate on catheter patency in patients older than 1 month. Finally, the study may be recreated using different catheter materials. The materials used in this study may affect the generalizability of the findings. Further, a similar study may be conducted using a small group of trained observers for the determination of specific complications related to the catheter. A recommendation would be to assess interrater reliability periodically.
References
Heparin in Peripheral Venous Infusion
American Academy of Pediatrics. (1974). Noise pollution: Neonatal aspects. Pediatrics, 54, 476. Banks, D. C., Cawdreg, H. M., Yates, D. B., Harris, M. G., & Kidner, P. H. (1970). Infection from intravenous catheters. Lancet, 1, 443-445. Batton, D. G., Maisels, M. J., & Applebaum, P. (1982).Use of peripheral intravenous cannulas in premature infants: A controlled study. Pediatrics, 70(3), 487-490. Birnbaum, D. W. (1981). Safety of maintaining IV sites longer than forty-eight hours. Journal of Clinical Microbiology, 13(5), 833-835. Bosque, E., & Weaver, L. (1987). Continuous versus intermittent heparin infusion of umbilical artery catheters in the newborn infant. Journal of Pediatrics, 208(1), 141143.
Brismar, B., Hardstedt, C., &Jacobson, S. (1982). Reduction of catheter-associated thrombosis in parenteral nutrition by intravenous heparin therapy. Archives of Surgery, 227, 1196-1199.
Collins, R. N., Braun, P. A., Zinner, S. H., & Kass, E. H. (1968).Risk of local and systemic infection with polyethylene intravenous catheters. New England Journal of Medicine, 279(7), 340-343. Coolinge, J. M., & Aranda, J . V. (1984). Nonmetabolic complications of neonatal intravenous therapy: Epidemiologic considerations. American Journal of Perinatology, 1, 185-189.
Daniell, H. W. (1973). Heparin in the prevention of infusion phlebitis. Journal of the American Medical Association, 226, 1317-1321.
David, R. J., Merten, D. F., Anderson, J. C., & Gross, S. (1981). Prevention of umbilical artery catheter clots with heparinized infusates. Developmental Pharmacology Therapeutics, 2, 117. Fabri, P. J., Mirtallo, R., & Ruberg, R. L. (1982). Incidence and prevention of thrombosis of the subclavian vein during total parenteral nutrition. Surgical Gynecology and Obstetrics, 255, 238-240. Fuchs, P. C. (1971). Indwelling intravenous polyethylene catheters: Factors influencing the risk of microbial colonization and sepsis. Journal of the American Medical Association, 226(9), 1447-1450. Gillhooly, J. T., Lindenberg, J. A,, & Reynolds, J. W. (1986). Survey of umbilical artery catheter practices. Clinical Research, 34, 142a. Gottfried, A. W., & Gaiter, J. L. (1985). Infants understress. Baltimore: University Park. Hoar, P. F., Wilson, R. M., Mangano, G. M., & Avery, G. J. (1981). Heparin bonding reducing thrombogenicity of pulmonary artery catheter. New EnglandJournal ofMedicine, 305(17), 993-995.
May/June 1992
Horgan, M. J., Bartoletti, A., Polansky, S., Peters, J. C., Manning, T. J., & Lamont, B. M. (1970). Effect of heparin infusates in umbilical artery catheters on frequency of thrombotic complications. Journal of Pediatrics, 111(5), 774-778.
Johnson, R. V., & Donn, S. M. (1988). Life span of intravenous cannulas in a neonatal intensive care unit. American Journal of Diseases of Children, 142, 968-971. Kitterman,J. A., Phibbs, R. H., & Tooley, W. H. (1970).Catheterization of umbilical vessels in newborn infants. Pediatric Clinics of North America, 17, 895. Long, J. G., Phillip, A. S., & Lucey, J. F. (1980). Excessive handling as a cause of hypoxemia. Pediatrics, 65, 203207.
Lowenbraun, S., Young, V., Kenton, D., & Serpick, A. (1970). Infection from intravenous scalp-vein needles in a susceptible population. Journal of the American Medical Association, 222(3), 451-453. Maki, D. G., Goldmann, D. A , , & Rhame, F. S. (1973). Infection control in intravenous therapy. Annals of Internal Medicine, 79, 867-887. Norris, S., Campbell, L. A., & Brenbert, S. (1982). Nursing procedures and alterations in transcutaneous oxygen tension in premature infants. Nursing Research, 32,330336.
Rajani, K., Goetman, B. W., Wenneberg, R. R., Turner, E., & Abilgaard, C. (1979). Effect of heparinization of fluids through an umbilical artery catheter on patency and frequency of complications. Pediatrics, 63(4), 375-378. Spiedel, B. D. (1982).Adverse effects of routine procedures on preterm infants. Lancet, I , 864-865. Tanner, W. A., Delaney, P. V., & Hennessy, T. P. (1980).The influence of heparin on intravenous infusions: A prospective study. BritishJournal of Surgery, 67, 311-312. Tulley, J. L., Friedland, G. H., Baldini, L. M., & Goldman, P. A. (1981). Complications of intravenous therapy with steel needles and Teflon catheters. American Journal of Medicine, 70, 702-706. Volpe, J. J. (1987). Neurology of the newborn (2nd ed.). Philadelphia: W. B. Saunders. Address for correspondence: Leslie Schaaf Treas, RNC, MSN, CNP, 12913 Goddard, Overland Park, KS 66213. Leslie Schaaf Treas is a neonatal nurse practitioner at Saint Luke’s Perinatal Center in Kansas City, Missouri. Barbara Latinis-Bridges is a professor of nursing at the University of Kansas in Kansas City. Dr. Latinis-Bridges is a member of NAA COG.
J O G N N
219
Emacy of Hepa& in Prnf@mal Vmous Infusion in Neonates Objective: To determine the eficacy of continuous, low dose heparin infusion in prolonging peripheral venous catheter patency in neonates. Design: Randomized, prospective study. Setting: Level III neonatal intensive-care unit. Participants: 113 neonates requiring IV therapy. Interventions: The neonates were randomly assigned to heparin (n = 63). The heparin group received '/z unit of heparin per milliliter of continuous intravenous infusate or intermittent heparin flush. Main outcome measures: Low-dose heparin infusion does not make a difference in the duration of peripheral venous catheter patency. The incidence of catheter-related complications in the heparin group is the same asfor those receiving no heparin. Results: The mean duration of catheter patency was 62.75 hours in the heparin group and 27.3 hours in the no-heparin group ( p = .0001). The occurrence of infection, bleeding, and extravastation injury was zero in the sample studied. The incidence of phlebitis was 18 cases in 132 (13%) in the heparfn group and 13 cases in 122 (10%) in the no-heparin group. According to chi-square analysis, the difference between groups was not statistically signgcant ( p > .05). Conclusion: Lowdose heparin infusion in peripheral venous catheters in neonates increased the duration of intravenous catheter patency ( p = .0001) without increased risks of bleeding, infection, phlebitis, or extravasation injury.
ecause nurses spend a significant amount of time inserting intravenous catheters for their patients, less frequent intravenous starts would allow nurses time to meet other needs of patients and their families. A more effective method of maintaining intravenous catheters in neonates is needed to reduce the identified stress factors associated with the insertion of intravenous catheters. Prolonging intravenous patency without increasing other risks to the newborn is beneficial to the patient and the caregiver.
Purpose Intravenous catheter insertion is time consuming and associated with significant risks to the neonate. Heparin has been used widely in adult patients to prolong the patency of the intravenous catheter, but few data exist regarding the use of low-dose heparin to maintain peripheral intravenous catheters in the neonate.
Intravenous Catheter insertion is time consuming and associated with significant risks to the neonate.
The purpose of this study was (a) to determine whether a low-dose heparin solution prolongs patency of the peripheral venous catheter in neonates and (b) to compare and contrast the nature and incidence of complications of infusion (e.g., phlebitis, infection, extravasation, bleeding) in neonates who receive low-dose heparin in their continuous intravenous infusates and those who do not.
Related Literature
Accepted: September 1991
Continuous low-dose heparin has been used widely to maintain the patency of central and peripheral intravenous catheters on the assumption that the low concentration of heparin accomplishes this goal without causing hemorrhagic complications (Rajani, Goetman, Wenneberg, Turner, & Abilgaard, 1979). Controlled studies in adults show the efficacy of heparin in reducing thrombophlebitis in peripheral (Gillhooly, Lindenberg, & Reynolds, 1986; Tanner, Delaney, & Hennessy, 1980) and central vein infusion sites (Brismar, Hardstedt, &Jacobson, 1982; Daniell, 1973; Fabri, Mirtallo, & Ruberg, 1982; Hoar, Wilson, Mangano, & Avery, 1981) and in prolonging the time until infiltration (Gillhooly et al., 1986).
214 J O G N N
Volume 21 Number 3
Heparin in Peripheral Venous Infusion
Heparin added to intravenous infusates more than doubled the duration of peripheral catheter patency without increasing risks to the neonate.
Studies in newborns show the benefits of both intermittent and continuous heparin infusion in preventing clot formation, reducing the incidence of line-related complications (Horgan et al., 1970), and prolonging the patency of umbilical vessel catheters without increasing the incidence of clinical bleeding (Bosque &Weaver, 1987; David, Merten, Anderson, & Gross, 1981; Gillhooly et al., 1986; Kitterman, Phibbs, & Tooley, 1981; Rajani et al., 1979). Few data exist, however, regarding the use of heparin in maintaining peripheral venous catheters in neonates. Alpan et al. (1984) conducted a controlled study of heparinization of total parenteral alimentation solutions administered through peripheral veins in premature infants. Heparin was found to double the duration of patency of intravenous catheters and to reduce significantly the incidence of phlebitis. The sample group, however, was small and incompletely controlled for several confounding variables relating to catheter life, the incidence of phlebitis, or both. N o data were found in the literature reviewing the use of heparin in peripheral dextrose and electrolyte solutions. Since this kind of fluid is commonly used in the sick neonate’s first days of life, the need to study this aspect of intravenous therapy was compelling.
Methods Subjects The convenience sample consisted of neonates in the level I11 neonatal intensive-care unit of a major teaching hospital. All of the neonates required peripheral intravenous therapy for routine medical care, regardless of birth weight or gestational age. Neonates were assigned to an experimental or control group based on the sequential number routinely assigned by the admitting office of the hospital. The control group included 49 neonates with 1 2 2 peripheral intravenous catheters to which no heparin was added. Saline was added to these intravenous catheters as an irrigating solution for intravenous insertion and intravenous locks. The experimental group included 63 neonates with 132 intravenous catheters to which one-half unit of heparin per milliliter of intravenous infusate and irrigating solution was added. All of the neonates in the study received intrave-
May/June 1992
nous therapy for varying periods and required a series of peripheral venous cannulations. Only the first three intravenous catheters per neonate were included in the study. Subjects excluded from the sample were those with (a) central line infusion (umbilical vessel or percutaneous central line), because heparin is routinely added to fluids administered through these catheters; (b) clinically suspected or documented bleeding diatheses; or (c) peripheral dopamine infusion. If a vasopressor was infused during the study, that intravenous catheter was eliminated from the data collection. ~~~~
Exclusion criteria included central line infusion, bleeding diathesis, and vasopressor use.
Procedure Registered nurses in the neonatal intensive-care unit who were assigned to the care of study subjects inserted the intravenous catheters. The nurses were randomly assigned to patients in both study groups. The intravenous insertion technique was consistent with unit protocol. The most suitable site available was permitted. The intravenous catheters were 24-gauge, % inch Insyte-W (Deseret Medical, Inc., Beckton & Dickinson Co., Sandy, UT). A clear, nonocclusive dressing (Op Site, Johnson and Johnson, New Brunswick, NJ, or Tegaderm, Medical Surgical Division, Saint Paul, MN) was used at the site to stabilize the intravenous catheter. N o antibiotic ointment or other dressing was used. The intravenous solution, in-line filter, and intravenous tubing were changed every 24 hours; medications were uniformly diluted, and blood products and medications were infused using a syringe infusion pump, according to unit protocol. The insertion site was inspected for complications at least every 2 hours by the registered nurse assigned to the neonate’s care. The intravenous line was removed if signs of local infection, phlebitis, or extravasation developed. The bedside or staff nurse recorded a description of the site when the line was removed. All nurses in the unit were instructed in how to define the characteristics of each intravenous complication before and intermittently during data collection. Moreover, a list of definitions of terms was provided to each nurse to ensure standard classification of intravenous site condition at the time of removal of the line. The definition of terms was located on each subject’s bedside clipboard and in various locations in the unit. Each nurse was trained to watch for signs of intravenous-related infection, extravasation, bleeding, and
J O G N N
215
C L I N I C A L
S T U D I E S
phlebitis. If signs of local infection developed, the cannula was cultured; if signs of local or systemic infection developed, two peripheral blood cultures were drawn from separate sites. The nurses were to observe for clinical signs of bleeding, i.e., prolonged bleeding from puncture sites and bloody endotracheal secretions. To detect intraventricular hemorrhage, routine cranial ultrasound was performed on all neonates born at or before 34 weeks’ gestation. All results were interpreted by the same radiologist, who had no knowledge of which group any given subject was assigned to.
Results The heparin and no-heparin groups shared several characteristics. The admission diagnoses of patients in both groups commonly included prematurity, multiple gestation, respiratory distress, suspected or documented sepsis, apnea, hyperbilirubinemia, anemia, and feeding intolerance. The birth weights and gestational ages of patients in the heparin and no-heparin groups were clinically comparable (see Table 1). The heparin additive was the only covariate that was statistically significant in influencing catheter patency.
The location of the intravenous catheter was comparable between the two groups. The hand or arm was the most common site for both groups. Other areas for catheter placement were similar between the groups (see Table 2). Infiltration was the most common reason for removing the intravenous catheter for both groups. Phlebitis was the second most common reason for discontinuing the catheters in the neonates (see Table 3).
Heparin (N
=
63)
Birth weight range Birth weight mean
8954,984 g 2,300 g
1,015-3,495 g
Gestational age range Gestational age mean
(SD, 806 g) 29-43 weeks 34.7 weeks (SD, 3.2 weeks)
(SD, 595 g) 28-42 weeks 33.8 weeks
216 J O G N N
Scalp Antecubital or saphenous Hand or arm Foot or leg
1,977 g
(SD, 3.0 weeks)
No heparin (N = 122)
Heparin (N = 132)
12 (9.84%) 12 (9.84%)
16 (12.12%)
73 (59.84%) 25 (20.49%)
24 (18.18%) 67 (50.76%) 25 (18.94%)
The no-heparin group received more medication than the heparin group (see Table 4). This difference is statistically significant ( p < .05). The heparin group received more total parenteral nutrition, more lipids, more calcium, and higher concentrations of glucose. Both groups received an equivalent number of blood transfusions (see Table 5). One reason for this study was to compare and contrast the duration of catheter patency between two groups of patients, one of which received low-dose heparin in the continuous intravenous fluid and the other of which received no added heparin in the infusate. The mean duration of patency for the 132 intravenous catheters in the heparin group was 62.75 hours; the mean duration of the 122 intravenous catheters in the no heparin group was 27.3 hours. Using the t test, this difference in duration of catheter patency was statistically significant ( p = .0001; t = 7.56). To determine whether the covariates (i.e., lipids, total parenteral nutrition, calcium, blood transfusion, glucose concentration, medications infused through the line) affected the duration of patency, the Cox proportional hazards model was calculated. This nonstepwise reTable 3. Comparison of Reason f o r Catheter Removal Between Groups Reason
Table 1. Birth Weight and Gestational Age by Group No heparin (N = 49)
Table 2. Comparison of Intravenous Sites Used Between Groups
Accidental dislodgement Elective discontinuation Infection Infiltration Leaking Mechanical Phlebitis Plugged
Volume 21 Number 3
No heparin (N = 122)
Heparin (N = 132)
5 (4.10%)
7 (5.30%)
8 (6.56%)
18 (13.64%)
0 (0)
86 (70.49%) 5 (4.10%) 2 (1.64%) 13 (10.66%) 3 (2.46%)
0 (0)
77 9 1 18 2
(58.33%) (6.82%) (0.76%) (13.64%) (1.52%)
Heparin in Peripheral Venous Infusion
Table 4. Comparison of Medications Infused Between Groups
Medicatton None Ampicillin and gentamicin Caffeine Ampicillin, gentamicin, and caffeine Ampicillin and caffeine Ampicillin Gentamicin Other
No hepartn (N = 122)
Heparin (N = 132)
56 (45.90%) 51 (41.80%)
88 (66.67%) 28 (21.21%)
3 (2.46%)
3 (2.27%)
3 (2.46%) 3 (2.46%) 5 (4.10%)
4 (3.03%)
0 (0) 1 (0.82%)
1 (0.76%)
1 (0.76%)
5 (3.79%) 2 (1.52%)
gression statistic was calculated to determine whether the covariates that were measured simultaneously affected the duration of catheter patency significantly. The heparin additive was the only covariate that influenced catheter patency significantly ( p < .OOOl). The study also examined catheter-related risks associated with the use of heparin. Chi-square analysis revealed no significant difference in the frequency of phlebitis in the first three intravenous catheters in either the heparin or the no-heparin group. The frequency of phlebitis in the no-heparin group was lo%, compared with 13%in the heparin group. The difference was not significant ( p > .05), according to Yate's correction for samples with 1 degree of freedom. Heparin infusion in the study was 0.8-3.2 units/ kg/hr. This low-dose infusion of heparin did not produce clinical symptoms of bleeding in the heparin group, and no incidence of bleeding was reported in the no-heparin group. The frequency of catheter-related infection in each group was zero, and the frequency of extravasation injury in each group was also zero.
tion of patency. The difference between the groups is significant ( p = .05). The groups received equivalent amounts of blood. Further research is needed to determine the effects of medication and infusates. The subjects had no infections during the study. Infection related to the catheter has complicated up to 8%of venous cannulations (Fuchs, 1971; Maki, Goldmann, & Rhame, 1973) and is said to result from many factors, including prolonged catheterization (Banks, Cawdreg, Yates, Harris, & Kidner, 1970; Birnbaum, 1981; Collins, Braun, Zinner, & Kass, 1968; Fuchs, 1971; Lowenbraun, Young, Kenton, & Serpick, 1970). Catheters that remained in situ for more than 72 hours in this study, however, caused no increased incidence of infection in neonates. In a study examining the life span of 5,000 intravenous catheters in neonates, the type of medication infused (except pancuronium bromide) did not have a discernible effect on the functional duration of patency (Johnson & Donn, 1988). None of the patients in this clinical study received pancuronium bromide. Phlebitis is known to complicate up to 18%of peripheral venous catheters (Batton, Maisels, & Applebaum, 1982; Tulley, Friedland, Baldini, & Goldman, 1981), but the frequency of phlebitis in this study sample was only 13% in the heparin group and 10%in the no-heparin group. The frequency of extravasation injury was zero. These findings of low frequencies of infection, extravasation, and phlebitis may be related to the Hawthorne effect. Since the nurses who inserted the intravenous catheters and assessed the intravenous site knew they were participating in an experimental study, they may have been more conscientious in their aseptic technique and assessment of intravenous sites to comply with the well-defined study protocol. Table 5. Variables That May Influence Duration of Catheter Patency
Discussion Vartable
The infusion of calcium, lipids, total parenteral nutrition, blood, medications, and high-concentration glucose solutions may shorten the catheter life of a peripheral intravenous line. In this study, the heparin group received more calcium, total parenteral nutrition, and lipids, as well as higher glucose loads, than did the no-heparin group. Although the heparin group received greater amounts of substances that may affect vascular integrity, the duration of patency more than doubled (see Table 4 ) . Conversely, the no-heparin group received more medications through the catheter, which may have been related to a shorter dura-
May/June 1992
No heparin (N = 49)
Total parenteral nutrition 1 (2.04%) Lipids 1 (2.04%) Glucose (high concentration) 2 (4.08%) Calcium 47 (95.92%) Blood 1 (2.04%) Medications 28 (57.14%)
Hepartn ( N = 63)
P
13 (20.63%) 0.003 8 (12.70%) 0.040 1 1 (17.46%) 0.028
63 (100%)
0.106 (ns)
0 (0) 0.255 (ns) 24 (38.10%) 0.045
Chi-square measurement with Yate's correction ns: not significant.
J O C N N
217
C L I N I C A L
S T U D I E S
The researcher was unable to control the activity level of the neonates, which could have an effect on the duration of catheter patency. The exclusion criterion of an umbilical catheter inherently limited how acute the study group’s medical condition would be. The researcher was unable to standardize the site used to one location. Staff nurses chose the most appropriate site for each neonate, and coincidentally, the intravenous sites of the groups were found to be similar.
Nursing Implications
Less frequent intravenous starts are benejicial in minimizing exogenous sources of stress that may lead to physiologic compromise of the sick neonate.
As the patient’s advocate, the nurse may promote the former method of intravenous therapy because less frequent intravenous starts may decrease neonatal and parental stress and lessen the potential for infection and other associated deleterious physiologic consequences of intravenous reinsertion. In addition, less frequent intravenous insertions may represent a cost savings to the family and the health-care facility because fewer materials for insertion and less nursing time would be needed.
The insertion of an intravenous catheter in a newborn often can be difficult because of a reduced number of available veins to cannulate. The procedure can be technically challenging to perform on small vessels, especially on neonates who are sick and have friable veins or poor perfusion. The insertion can cause pain, which can lead to deterioration in the neonate’s condition. Furthermore, the pain response associated with intravenous cannula reinsertion may be associated with a transient rise in the neonate’s blood pressure, a known risk factor for intraventricular hemorrhage (Volpe, 1987). These sick neonates have a diminished capacity to adapt to exogenous stressors in the environment, including noise, temperature fluctuations with handling, and procedures. A sick neonate’s response to overstimulation is manifested by apnea and bradycardia, cyanosis, hypertension and hypotension, hypoxemia, hyperglycemia, elevation of intracranial pressure, and feeding difficulties (American Academy of Pediatrics, 1974; Coolinge & Aranda, 1984; Gottfried & Gaiter, 1985; Long, Phillip, & Lucey, 1980; Norris, Campbell, & Brenbert, 1982; Spiedel, 1982). Complications of intravenous catheter insertion in neonates include (a) hypothermia due to exposure to ambient room temperatures during the insertion of the intravenous catheter, (b) hypoglycemia related to the interruption of glucose infusion, (c) sleep deprivation and hypoxia from the stress of insertion, and (d) increased infection potential with repeated venipuncture (Long et al., 1980; Norris et al., 1982; Spiedel, 1982). In caring for the compromised neonate, the neonatal nurse attempts to minimize exogenous sources of stress and prevent further physiologic compromise. Less frequent restarts of the intravenous infusion are beneficial. Continuous intravenous infusion with heparin added in the peripheral venous catheter prolonged the patency of the intravenous line an average of 62.7 hours. Solutions with no heparin added in the infusate remained patent for an average of 27.3 hours.
Alpan, G., Eyal, F., Springer, G., Glick, B., Groder, K., & Armon, L. (1984). Heparinization of alimentation solutions: A controlled study. Pediatrics, 74, 375-378.
218 J O G ”
Volume 21 Number 3
Recommendations f o r Future Research Replication of this scientific study is needed to validate or dispute its findings. The use of heparin solution versus saline flush for the irrigation of intravenous locks in neonates may be compared. Conflicting data are reported in the literature about adult subjects and the efficacy of heparin irrigation solution in maintaining the patency of intravenous locks. The study could be replicated by comparing the patency of the intravenous catheter without heparin in the infusate to the patency of catheters with the infusion of 1 unit per milliliter of solution. The dosage used in this study examined the effect of ‘/z unit per milliliter of infusate. In many neonatal intensive-care units nationwide, however, the addition of 1 unit of heparin per milliliter is standard. The effect of an increased dose of heparin may influence the duration of patency. The study could be replicated by using a pediatric sample. Few data exist regarding the use of continuous heparin infusate on catheter patency in patients older than 1 month. Finally, the study may be recreated using different catheter materials. The materials used in this study may affect the generalizability of the findings. Further, a similar study may be conducted using a small group of trained observers for the determination of specific complications related to the catheter. A recommendation would be to assess interrater reliability periodically.
References
Heparin in Peripheral Venous Infusion
American Academy of Pediatrics. (1974). Noise pollution: Neonatal aspects. Pediatrics, 54, 476. Banks, D. C., Cawdreg, H. M., Yates, D. B., Harris, M. G., & Kidner, P. H. (1970). Infection from intravenous catheters. Lancet, 1, 443-445. Batton, D. G., Maisels, M. J., & Applebaum, P. (1982).Use of peripheral intravenous cannulas in premature infants: A controlled study. Pediatrics, 70(3), 487-490. Birnbaum, D. W. (1981). Safety of maintaining IV sites longer than forty-eight hours. Journal of Clinical Microbiology, 13(5), 833-835. Bosque, E., & Weaver, L. (1987). Continuous versus intermittent heparin infusion of umbilical artery catheters in the newborn infant. Journal of Pediatrics, 208(1), 141143.
Brismar, B., Hardstedt, C., &Jacobson, S. (1982). Reduction of catheter-associated thrombosis in parenteral nutrition by intravenous heparin therapy. Archives of Surgery, 227, 1196-1199.
Collins, R. N., Braun, P. A., Zinner, S. H., & Kass, E. H. (1968).Risk of local and systemic infection with polyethylene intravenous catheters. New England Journal of Medicine, 279(7), 340-343. Coolinge, J. M., & Aranda, J . V. (1984). Nonmetabolic complications of neonatal intravenous therapy: Epidemiologic considerations. American Journal of Perinatology, 1, 185-189.
Daniell, H. W. (1973). Heparin in the prevention of infusion phlebitis. Journal of the American Medical Association, 226, 1317-1321.
David, R. J., Merten, D. F., Anderson, J. C., & Gross, S. (1981). Prevention of umbilical artery catheter clots with heparinized infusates. Developmental Pharmacology Therapeutics, 2, 117. Fabri, P. J., Mirtallo, R., & Ruberg, R. L. (1982). Incidence and prevention of thrombosis of the subclavian vein during total parenteral nutrition. Surgical Gynecology and Obstetrics, 255, 238-240. Fuchs, P. C. (1971). Indwelling intravenous polyethylene catheters: Factors influencing the risk of microbial colonization and sepsis. Journal of the American Medical Association, 226(9), 1447-1450. Gillhooly, J. T., Lindenberg, J. A,, & Reynolds, J. W. (1986). Survey of umbilical artery catheter practices. Clinical Research, 34, 142a. Gottfried, A. W., & Gaiter, J. L. (1985). Infants understress. Baltimore: University Park. Hoar, P. F., Wilson, R. M., Mangano, G. M., & Avery, G. J. (1981). Heparin bonding reducing thrombogenicity of pulmonary artery catheter. New EnglandJournal ofMedicine, 305(17), 993-995.
May/June 1992
Horgan, M. J., Bartoletti, A., Polansky, S., Peters, J. C., Manning, T. J., & Lamont, B. M. (1970). Effect of heparin infusates in umbilical artery catheters on frequency of thrombotic complications. Journal of Pediatrics, 111(5), 774-778.
Johnson, R. V., & Donn, S. M. (1988). Life span of intravenous cannulas in a neonatal intensive care unit. American Journal of Diseases of Children, 142, 968-971. Kitterman,J. A., Phibbs, R. H., & Tooley, W. H. (1970).Catheterization of umbilical vessels in newborn infants. Pediatric Clinics of North America, 17, 895. Long, J. G., Phillip, A. S., & Lucey, J. F. (1980). Excessive handling as a cause of hypoxemia. Pediatrics, 65, 203207.
Lowenbraun, S., Young, V., Kenton, D., & Serpick, A. (1970). Infection from intravenous scalp-vein needles in a susceptible population. Journal of the American Medical Association, 222(3), 451-453. Maki, D. G., Goldmann, D. A , , & Rhame, F. S. (1973). Infection control in intravenous therapy. Annals of Internal Medicine, 79, 867-887. Norris, S., Campbell, L. A., & Brenbert, S. (1982). Nursing procedures and alterations in transcutaneous oxygen tension in premature infants. Nursing Research, 32,330336.
Rajani, K., Goetman, B. W., Wenneberg, R. R., Turner, E., & Abilgaard, C. (1979). Effect of heparinization of fluids through an umbilical artery catheter on patency and frequency of complications. Pediatrics, 63(4), 375-378. Spiedel, B. D. (1982).Adverse effects of routine procedures on preterm infants. Lancet, I , 864-865. Tanner, W. A., Delaney, P. V., & Hennessy, T. P. (1980).The influence of heparin on intravenous infusions: A prospective study. BritishJournal of Surgery, 67, 311-312. Tulley, J. L., Friedland, G. H., Baldini, L. M., & Goldman, P. A. (1981). Complications of intravenous therapy with steel needles and Teflon catheters. American Journal of Medicine, 70, 702-706. Volpe, J. J. (1987). Neurology of the newborn (2nd ed.). Philadelphia: W. B. Saunders. Address for correspondence: Leslie Schaaf Treas, RNC, MSN, CNP, 12913 Goddard, Overland Park, KS 66213. Leslie Schaaf Treas is a neonatal nurse practitioner at Saint Luke’s Perinatal Center in Kansas City, Missouri. Barbara Latinis-Bridges is a professor of nursing at the University of Kansas in Kansas City. Dr. Latinis-Bridges is a member of NAA COG.
J O G N N
219
