Blood Clots in Intravenous Catheters with Positive Pressure Injection Caps
 by Guy D. LaRue, CRNI

Abstract: Mild resistance has been noted when flushing some patient’s peripheral intravenous catheters with normal saline via positive pressure injection caps.  The hypothesis was that, despite the initial positive pressure produced by these injection caps, blood entered the catheters, formed a thrombus and was embolized into the blood stream upon flushing. A double-blinded, randomized study was conducted to examine the contents of catheters with positive pressure injection caps flushed with either heparin or normal saline.  The results revealed that all of the study catheters contained blood regardless of flush solution.  The saline flushed catheters had an average of 14 times larger clots than heparin flushed catheters.  A scientific rational and the related healthcare issues are discussed.


It has been proposed that simply filling intravenous catheters with fluid will prevent blood from entering that catheter.1 The rationale for this assumption appears to be that once a fluid fills a space, no other fluid can enter that space.2 Since the late 1970’s a dilute solution of heparin was used to fill intermittently used peripheral intravenous catheters.3 This was done to prevent coagulation of blood in the catheters in the event that blood entered the catheter.3  Many researchers found that a dilute solution of heparin in normal saline was more effective than normal saline at preventing catheter occlusions.4 5 6 7 However, other research concluded that both saline and heparin flush solutions were equally effective at ensuring catheter patency.1 8 9 10 11 12 13 14 15 16 As a result of the latter research, in 1998 the Intravenous Nurse Society changed its standard of practice for flushing peripheral intravenous catheters from heparin solution to normal saline solution.17

In the early 1990’s needleless injection caps were introduced in intravenous therapy to reduce the potential for injury to clinicians from contaminated metal needles.18 These early needleless cap systems caused inadvertent aspiration of blood into catheters during disconnect procedures.19 Aspiration of blood into intravenous catheters during disconnect procedures resulted from a negative pressure (or vacuum) equal to the volume of the portion of the device being pulled out of these injection caps.20 Medical supply manufacturers eventually designed catheter injection caps that do not result in negative pressure and do not aspirate blood during disconnect procedures.21 22

Currently there are two basic types of needleless injection caps that are designed to not aspirate blood during disconnect procedures.  One type of injection cap is referred to as a neutral pressure injection cap, and the other is referred to as a positive pressure injection cap. In vitro demonstrations have demonstrated that both types of injection caps are essentially effective at preventing aspiration of external fluid into the intravenous catheter by maintaining either a neutral pressure or by creating a small, momentary positive pressure within the catheter during disconnect procedures.  

Advertisements for positive pressure injection caps have claimed that the caps are the most reliable way to help maintain the patency of IV catheters, and that no additional steps need to be taken to prevent backflow.23 24 Another advertisement claims that the cap prevents the retrograde flow of blood into the catheter lumen.25 Some authors, without citing either manufacturer claims or research references, have written that specific injection caps eliminate the need for heparin as they prevent blood from entering peripheral intravenous catheters.26 27

Several nursing peers reported to the primary author that they regularly experienced minor resistance when flushing some peripheral intravenous catheters (with and without neutral or positive pressure injection caps) followed by a sudden sensation of easy flushing. This phenomenon was noted particularly since the change in flushing standards from heparin to normal saline flush solutions. The primary author suspected the existence of blood clots in these catheters that were being flushed into the patients circulation with mild syringe pressure. It was hypothesized, regardless of the above noted claims by manufacturers, that clots have been forming in intravenous catheters equipped with these special injection caps.  


 It was decided to perform a study of catheters with positive pressure injection caps flushed with either heparin or normal saline that measured the pressure required to flush the catheters and provided a thorough examination of the catheter contents.

A prospective, double-blinded and randomized study was designed to analyze the contents of short peripheral intravenous catheters with attached positive pressure injection caps eight hours after being flushed with either normal saline or heparin solutions.  All catheters were placed in a study subject who had pre-study lab work performed to rule out blood anomalies. Catheters were placed in similar veins of each arm during each trial. For example; if one catheter was placed in a vein on the posterior medial aspect of the left hand then the second catheter of that trial was placed in the vein on the posterior medial aspect of the right hand.  Two, one-inch 20 gauge same-brand intravenous catheters, with same-brand 5 inch extension tubing and with same-brand positive pressure injection caps, flushed with 3 mls of 10 Units/ml heparinized saline or 3 mls of NaCl 0.9% were used in each trial.  

Normal patient activity was simulated during the main study consisting of 20 trials.The catheters were placed into the study subject prior to preparing for bed. The study subject then brushed his teeth, washed his hands and slept about 7 hours.  Approximately 8 hours after insertion, the catheters were removed, attached to a 3 ml syringe with a pressure gauge and flushed with 1 ml of normal saline directly onto an electronic weight scale. The amount of pressure required to flush the catheter contents onto the electronic scale plate was measured in pounds per square inch (psi). The saline solution and liquid blood was carefully wicked off the electronic scale plate with tissue paper. The remaining clot material was weighed. The amount of pressure required to flush each catheter and the weight of each clot was recorded.

A pilot of three trials was performed to assess the study methodology.  During the pilot an unexpected finding impacted the main study.  It was noted that the catheters and extension tubing did not appear to have blood in them as long as the cannulated hands or arms were in a level or elevated position.  When the cannulated hands or arms were lowered to dependent positions, blood was seen falling down into the clear extension tubing of some of the catheters.  This finding indicated that movement and catheter position had impact on blood entering the pilot study catheters. As a result, arm movement and catheter movement were added to the variables controlled for in the main study.  The study subject’s activities attempted to mimic activities that a typical hospital patient might perform.  The results of the pilot study remained double-blinded until the main study was complete so as not to bias the outcome.



Clot Analysis

  Average PSI* Range Of PSI Average Mass^ Range of Mass
Saline 3.3 0 - 8 0.0145 0 - 0.03
Heparin 0.4 0 - 4 0.001 0 -0.01
* pounds per square inch
^ grams

Table 1

The average syringe pressure required to flush the saline flushed catheters was eight times greater than the pressure required to flush the lines flushed with heparin.  Even with this increase in resistance to flush, the catheters posed no difficulty to flush manually with the 5 ml syringes used in the experiment.  As hypothesized, the catheters with increased resistance to thumb pressure and then a sudden lack of resistance did in fact contain some of the largest clots in this study.

Every catheter in the study had blood flushed out of it onto the electronic scale plate.  Only a few of the heparin-flushed catheters had visible clots after the liquid  was wicked away (see photographs 1 and 2 for examples of typical study catheter contents immediately after being removed from the study subject and flushed onto the weight scale plate).

Photograph 1 shows contents of a typical catheter flushed eight hours prior with 3 mls of normal saline. This shows the main clot prior to wicking the excess liquid away for weighing. Photograph 2 shows contents of a typical catheter flushed
eight hours prior with 3 mls of heparin, 10 units per ml. In this
example, once the liquid was wicked away there was no measurable clot.

The total clot mass in the 20 saline-flushed catheters was 0.29 grams or 14.5 times greater than the total clot mass of 0.02 grams in the heparinized catheters. The range of clot mass in the normal saline flushed catheters was 0 to 0.03 grams.  The range of clot mass found in the heparin flushed catheters was 0 to 0.01 grams. (See Table 1) Correlation is significant at the 0.01 level (1-tailed).

One unexpected finding in this study was the presence of clotted blood adhering to the inside of catheters flushed with normal saline.  Photograph 3 illustrates the presence of clots found inside some of the junctions of the catheters and the extension tubings after the main clots had been flushed clear.  These clots were not discovered until near the end of the study and as a result were not measured nor added to the clot weight data.

Photograph 4 shows the typical external appearance of both a heparin and saline flushed catheter after having their contents flushed onto the weight scale plate.  Note the darker color of the white cannula and the visible blood attached to the hub section as well as to the extension tubing in the catheter that had been flushed eight hours earlier with normal saline. Compare that to the clean appearance of the catheter and extension tubing of the catheter flushed with heparin eight hours earlier.


Photograph 3. The white arrow indicates a blood clot found inside the junction of the catheter hub and the extension tubing after
being supposedly flushed clear. The small round clot between the hub and the extension tubing in Photograph 1 is also from inside the junction.
Photograph 4. Two catheters from the same trial are shown
against a white background alter being flushed "clear" of blood.
The upper catheter was flushed eight hours earlier with normal saline.
The bottom catheter was flushed eight hours earlier with heparin.


Several conclusions were drawn from this study:

1.Neither fluid in intravenous catheters nor positive pressure injection caps prevents blood from entering open-ended catheters. While many causes for blood entering intravenous catheters have been suggested, this study finds that gravity, while not previously suggested, is a principle reason for blood found in intravenous catheters.19 25  Fluids with greater densities are heavier than fluids with lesser densities. When a fluid with greater density is added to a vial filled with a less dense fluid, the greater density fluid will fall through and displace the less dense fluid.28

Blood has a specific gravity of approximately 1.05 which means it is denser than normal saline at a specific gravity of 1.02.29. When added to a container of normal saline solution, blood will sink to the bottom of that container. This experiment shows that blood will also fall into an open-ended venous catheter if the catheter opening is directed upwards.  As the blood falls into the catheter, it displaces the lighter flush solution.

This phenomenon was demonstrated as part of a poster presentation at the January 2006 Association for Vascular Access conference in Savannah, Georgia. The author photographed an easy-to-replicate, in vitro demonstration of this phenomenon using a 3 millimeter diameter glass tube with a standard neutral pressure injection cap attached to one end and a nearly full vial of citrated human blood attached to the other end.  The cap and tubing were flushed with normal saline while being held parallel to the ground.  No blood was observed in the glass tubing for several minutes. As soon as the vial was held in an inverted position (superior to the saline flushed tubing), blood was visible falling down through the tubing.  Time-lapsed photographs 5 through 7 show blood filling the glass tube and displacing the normal saline flush solution.


Photographs 5, 6 and 7 show blood falling into a glass tube when the vial of blood is held above the saline-flushed tube.

The pictures are taken approximately 20 seconds apart with photograph 5 taken first and photograph 7 taken last.

2. Conclusions reached by past studies that relied on the subjective ability of clinicians to determine catheter patency by their ability to push a syringe plunger with their thumb are suspect.1 9 10 12 14 15 16 Many of the studies that found no difference between heparin and saline flushes had the initial appearance of being highly reliable by virtue of being  prospective, randomized and double-blinded with large samples of patients. The single problem with those studies is that they failed to provide accurate construct validity which refers to the extent a given research measure (the clinicians’ ability to flush catheters) depicts the concept of interest (catheter patency).30  In other words, clinicians’ ability to flush the experiment catheters could not accurately determine whether or not there were clots in those same catheters. Other health care professionals have identified this particular flaw in catheter patency research.31 32 33

3. It is probable that clots similar to the clots found in this study have been flushed into patients on a regular basis since the standards of practice for flushing peripheral intravenous catheters changed in the mid and late 1990’s.17 34  It has been proposed that clots from intravenous catheters flushed into patient’s bloodstreams are potentially life-threatening.31  If there has been no increase in reports of pulmonary embolisms since the flushing standards change, then that alone is a remarkable finding.  With the numbers of patients in this inadvertent but very large scale public experiment it may be safe to conclude that clots of this size are not as dangerous as previously asserted.

4. This study found that blood remains in saline-flushed catheters even after the main clots are flushed clear. Based on this study finding alone, the standard of practice for flushing peripheral intravenous catheters should be reviewed with a bias toward a return to low dose heparin or the equivalent. Blood clots within catheter lumens may lead to catheter related infections.35 36 37  Catheter related infections promote the increased use of antibiotics with the resultant increased resistance to those antibiotics.38  Intravenous catheter infections increase the rate of patient morbidity and mortality.39  Use of heparin in intravenous catheters has been shown to decrease the incidence of catheter related sepsis. Of further clinical importance is that the use of saline flushes not be extrapolated to other types of vascular access devices.   


This study was a crossover case/control. The research subject served as their own control therefore limiting confounding variables in the experiment. During the blinded peer-review process, some reviewers asserted that this experiment has a sample size of one (the human subject) rather than a sample size of 40 (the number of devices and flush samples) claimed in this paper. This is a specious argument. The reviewers are comparing this experiment with standard trials where a single medication (or device) is tested to determine the effects on the animal or human subjects. In this experiment, the effect of the devices on the human subject was of no consequence to the phenomenon being researched. The sole purpose of this experiment was to determine the effects of normal blood on intravenous catheter devices. If the twenty trials in this experiment had been done via a bench test (similar to the bench test shown in photographs 5, 6 and 7) or on a single animal subject it is doubtful the reviewers would have been confused by the sample size. It was decided during the design phase of this research that the results of the experiment would have greatest validity if tested in a controlled venous system rather than on a single bench test device. The primary reason for a human subject (as opposed to a bench test) was that comparing heparin with saline flushes is not possible due to clotting that occurs once untreated blood is removed from a vascular system. The author believes that the findings in this article are important regardless of the small size of the samples (twenty comparative flush trials, forty devices and a single bench (albeit human) test device). The author feels that it would be a mistake to minimize the finding of this small study and feels highly confident that future research that replicates the basic methodology described in this article will find similar phenomenon.

This study was initiated in early 2001 and the findings were first introduced via a poster presentation and a lecture at the 19th Annual Association for Vascular Access Conference in September, 2006.41 By the time this article was submitted for publication in late 2007, publication standards for two primary intravenous therapy journals had changed such that any research that involved human subjects required formal Investigation Review Board (IRB) approval. At the time this study was initiated this requirement was not known to the primary author who had previously published and presented without specific IBR approval.41 42 43  While an IRB review was not obtained for this research, the principles of the Declaration of Helsinki as developed by the World Medical Association at the time were adhered to throughout this research project. The subject was fully aware of the potential benefits and risks of participating in this study. It was decided by the author that the findings of this study have such high potential for improving outcomes for patients that this article must be published regardless of current journal policies. The decision was made to publish the paper independently via the internet.  

In order to validate the theory presented in this article that blood density is a primary cause of blood entering catheters more studies of a similar nature need to be done. A suggestion for doing a similar experiment would be to simply remove and examine the contents of intravenous catheters removed under normal clinical circumstances of large numbers of active patients who have received blinded flushes of either saline or heparinized saline.  

Future research on the effectiveness of all solutions and devices designated to keep intravenous catheters patent should include careful analysis and measurement of catheter contents on removal.  Future research on the effectiveness of solutions and devices designed to keep intravenous catheters patent should also include documenting patient activity and catheter positions to account for the potential effects of gravity on blood.

This study was limited to testing one concentration of heparin.  In light of the potential complications with chronic, high-dose heparin use, other compounds/solutions/chemicals should continue to be studied to prevent clot formation in intravenous catheters. We suggest different strengths of heparin be tested to find the lowest dose that prevents clot formation and minimizes the risk of adverse effects.

The author suggests that future research on devices purported to prevent blood out of intravascular catheters have particular internal parts of the control group devices removed to render their effect neutral. This would facilitate randomized, double-blinded studies of these devices and provide valid research related to these products.


1 Shearer J. Normal saline flush versus dilute heparin flush; a study of peripheral intermittent IV devices. NITA. 1989;Nov/Dec:425-427.

2 Schustek M. A cost-effective approach to PRN device maintenance. NITA. 1984;Nov/Dec:527.

3 Augspuger EF, Davis LF. Heparin lock intravenous technique. J Oral Surg. 1974;32:786.

4 Hanson RL, Grant AM, Majors KR. Heparin-lock maintenance with ten units of sodium heparin in one      milliliter of normal saline solution. Surg Gynecol Obstet. 1976;142:373-376.

5 Meyer BA, Little CJ, Thorp JA, Cohen GR, Yeast JD. Heparin versus normal saline as a peripheral line flush in maintenance of intermittent intravenous lines in obstetric patients. Obstetrics and Gynecol. 1995;85(3):432-436.

6 Mudge B, Forcier D, Slattery MJ. Patency of 24-gauge peripheral intermittent infusion devices: A comparison of heparin and saline flush solutions. Pediatric Nurs. 1998;24(2):142-149.

7 Danek GD, Noris EM. Pediatric IV catheters: efficacy of saline flush. Pediatric Nurs. 1992;18(2):111-113.

8 Hamilton RA, Plis JM, Clay C, Sylvan L. Heparin sodium versus 0.9% sodium chloride injection for maintaining patency of indwelling intermittent infusion devices. Clinical Pharm. 1988;7:439-443.

9 Lombardi TP, Gunderson B, Zammett LO, Walters JK, Morris BA. Efficacy of 0.9% sodium chloride injection with or without heparin sodium for maintaining patency of intravenous catheters in children. Clinical Pharm. 1988;7:832-836.

10 Taylor N, Hutchinson E, Milliken W, Larson E. Comparison of normal versus heparinized saline for flushing infusion devices. J Nurs Qual Assurance. 1989;3(4):49-55

11 Garrelts JC, LaRocca J, Ast D, Smith Jr DF, Sweet DE. Comparison of heparin and 0.9% sodium chloride injecton in the maintenance of indwelling intermittent IV devices. Clinical Pharm. 1989;8:34-39.

12 Barrett PJ, Lester RL. Heparin versus saline flushing solutions in a small community hospital. Hosp Pharm. 1990;25:115-118.

13 Tuten SH, Gueldner SH. Efficacy of sodium chloride versus dilute heparin for maintenance of peripheral intermittent intravenous devices. Appl Nurs Research. 1991;4(2):63-71.

14 Geritz MA. Saline versus heparin in intermittent infuser patency maintenance. West J Nurs Research. 1992;14(2):131-141.

15 Kleiber C, Hanrahan K, Fagan CL, Zittergruen MA. Heparin vs saline for peripheral IV locks in children. Ped Nurs. 1993;19(4):405-409. 

16 Epperson EL. Efficacy of 0.9% sodium chloride injection with and without heparin for maintaining indwelling intermittent injection sites. Clinical Pharm. 1984;3:626-629.

17 Journal of Intravenous Nursing Intravenous Revised Intravenous Nursing Standards of Practice. 1998;21(1S):61-62.

18 Hedrick C. A process for selecting a safety needle system in a community teaching hospital. J Intravenous Nurs. 1993;16(5):299-302.

19 Krzywda EA. Predisposing factors, prevention and management of central venous catheter occlusions. J Intravenous Nurs. 1999;22(6S):11-17.

20 Mayo DJ. Reflux a manageable problem. J Vascular Access Dev. 2001;6(4):39-40.

21 Lenhart C. Prevention vs treatment of VAD occlusions. J Vascular Access Dev. 2000;5(4):34-35.

22 Berger L. The effects of positive pressure devices on catheter occlusions. J Vascular Access Dev. 2000;5(4):31-33.

23 Advertisement for CLC2000 Catheter Patency Device, ICU Medical, Inc., San Clemente, CA. J Vascular Access Dev. 2005;10(3):inside front cover.

24 Pamphlet advertisement for the Becton Dickenson Posiflow IV Access System, Becton Dickenson and Company, Franklin Lakes, New Jersey.

25 Advertisement for CLC2000, ICU Medical, Inc., San Clemente, CA. J Intravenous Nurs. 1998;21(6):inside back cover.

26 Goode CJ, Titler M, Rakel B, et al. A meta-analysis of effects of heparin flush and saline flush: quality and cost implications. Nurs Research. 1991;40(6):324-330.

27 Hadaway L. Heparin locking for central venous catheters. J Assoc for Vascular Access. 2006;12(3):224-231.

28 Knight RD. Physics for Scientists and Engineers; A Strategic Approach. San Francisco, CA.: Addison Wesley. 2004;pp.460.

29 Tortora GJ, Anagnostakos NP. Principles of Anatomy and Physiology. San Francisco, CA.: Canfield Press. 1978;pp.424

30 Norwood SL. Research Strategies for Advanced Practice Nurses. Upper Saddle River, NJ.: Prentice-Hall. 2001;pp.14

31 Ferraro JM, Guanci RF, Stoneburner RJ. Letter to the editor. Clinical Pharm. 1985;4(5):488-489.

32 Jordan L. Should saline be used to maintain heparin locks? Focus on Crit Care. 1991;18:144-150

33 Gardner C. Normal saline flush vs dilute heparin flush. J Intravenous Nurs. 1989;11(3):206.

34 American Society of Hospital Pharmacists. ASHP therapeutic position statement on the institutional use of 0.9% sodium chloride injection to maintain patency of peripheral indwelling intermittent infusion devices. Am J Hosp Pharm. 1994;51:1572-1574.

35 Timsit JK, Farkas JC, Boyer JM, et al. Central vein catheter-related thrombosis in intensive care patients: incidence, risk factors, and relationship with catheter-related sepsis. Chest. 1998;114:207-213.

36 Jurado R, Ribeiro M. Possible role of systemic inflammatory reaction in vascular access thrombosis. South Med J. 1999;92:877-881.

37 Vaudaux P, Pittet D, Haeberli A, et al. Host factors selectively increase staphylococcal adherence on inserted catheters: a role for fibronectin and fibrinogen or fibrin. J Infectious Diseases. 1989;160(5):856-878.

38 Carrico RM, Arnold F, Goss LK. Antimicrobial resistance. J Infusion Nurs. 2005;28(3):183-187.

39 Jones GR. A practical guide to evaluation and treatment of infections in patient with central venous catheters. J Intravenous Nurs. 1998;21(S5):134-142.

40 Bailey J. Reduction of catheter-associated sepsis in parenteral nutrition using low-dose intravenous heparin. British Med J. 1979;1:1671-1673.  

41 LaRue GD, Kondo LM. Blood in Intravenous Catheters with Positive Pressure Injection Caps. 19th Annual AVA Conference handbook, 2006:189.

42 LaRue GD. Improving Central Placement Rates of Peripherally Inserted Catheters. J Intravenous Nurs. 1995;18(1):24-27.

43 LaRue GD.  Efficacy of Ultrasonography in Peripheral Venous Cannulation. J Intravenous Nurs. 2000; 23(1):29-34.



Home | About Us | Published Articles | Recent Articles | Topics Of Interest

Vascular Research Association
7827 20th Dr NE
Marysville, WA 98271