Current issues of ACP Journal Club are published in Annals of Internal Medicine


Silicone catheters led to fewer episodes of thrombophlebitis than did teflon cannulas in peripheral intravenous nutrition

ACP J Club. 1992 May-June;116:77. doi:10.7326/ACPJC-1992-116-3-077

Source Citation

Madan M, Alexander DJ, McMahon MJ. Influence of catheter type on occurrence of thrombophlebitis during peripheral intravenous nutrition. Lancet. 1992 Jan 11; 339:101-3.



To compare the occurrence of thrombophlebitis with the use of fine-bore silicone or larger teflon catheters for peripheral intravenous nutrition (PIN).


Randomized controlled trial.


Nutritional support service of an English general hospital.


50 of 165 patients referred to the service were thought to require nutritional support for < 10 days and had suitable peripheral veins for PIN. The reasons for referral were gastrointestinal tract surgery (47 patients) and preoperative preparation (3 patients). 27 patients aged 29 to 78 years (median 64 years) were randomized to the silicone-catheter group and 23 patients aged 17 to 87 years (median 74 years) to the teflon-cannula group.


The 23-gauge silicone rubber catheter had an outer diameter of 0.6 mm by 15 cm long; the 20-gauge teflon cannula was 1.0 mm by 3.2 cm. All tubes were inserted, using aseptic procedures, into the patient's largest vein palpable in the proximal third of the forearm. Thus, for the silicone catheter, the tip was close to the antecubital fossa. The nutrient solution (osmolality, 1250 mOsm/kg) was delivered by infusion pump at 110 mL/h. Each 2.5-L infusion bag contained 1800 kcal; heparin, 500 U/L, and hydrocortisone, 5 mg/L, were added. A patch, providing 5 mg of glyceryl trinitrate in 24 hours, was placed distal to the insertion site and changed every 48 hours.

Main outcome measures

Thrombophlebitis was defined as the presence of a raised erythematous area 0.5 C degrees warmer than the surrounding skin. Secondary measures were the removal of catheters for reasons other than thrombophlebitis and the length of time catheters stayed in place.

Main results

The groups were fed intravenously for similar periods (median 5 d). Thrombophlebitis developed in 2 of 27 patients (7%) in the silicone-catheter group compared with all 23 patients (100%) in the teflon-cannula group {P < 0.001. This absolute risk reduction of 93% means that 2 patients would need to be fed via a silicone catheter (compared with teflon cannula) to prevent 1 additional patient from developing thrombophlebitis, 95% CI 1 to 2; the relative risk reduction was 93%, CI 76% to 98%}.* 4 occluded silicone catheters were removed compared with 20 teflon cannulas that were removed because of extravasation, edema, or pain without thrombophlebitis. The silicone catheters functioned longer than the teflon cannulas (median 128 h vs 36 h, P < 0.001).


Peripheral intravenous nutrition through a silicone catheter led to fewer patients developing thrombophlebitis than did feeding via teflon cannula.

Source of funding: Not stated.

Address for article reprint: Mr. M.J. McMahon, Nutritional Support Service, Department of Surgery, General Infirmary, Leeds LS1 3EX, United Kingdom.

*Numbers calculated from data in article.


Central venous nutrition (the most common method of TPN or "IV hyperalimentation") can deliver large volumes of hypertonic nutrient solution over prolonged periods; the disadvantages include an infection rate of ≥ 4% and complex plumbing. The disadvantage of peripheral venous access is the high incidence of phlebothrombosis and thrombophlebitis. Techniques to protect PIN include limiting osmolality to ≤ 600 mOsm/kg—requiring large volumes of fluid—or incomplete provision of nutrients. The authors have pulled all stops in facilitating a trial of PIN by the addition of heparin and hydrocortisone with a vasodilator patch to maintain blood flow and thus avoid thrombosis. In this study, as in an earlier animal model study (1), a fine silicone catheter was impressively safer and more effective than teflon. The two catheter types, however, also differed in diameter (23 gauge vs 20 gauge); in length (15 cm vs 3.2 cm); and in the technique of catheter insertion through or over the needle. Meticulous care with both catheters allowed the infusion of hypertonic (1250 mOsm/kg) solution with only 2 catheter changes in the silicone group during a median 5-day infusion. Even with multiple catheter changes, however, the teflon sleeves were much cheaper.

Practitioners of parenteral nutrition may replace standard teflon sleeves with silicone catheters in selected high-risk patients or in those with poor peripheral veins. Total parenteral nutrition via a central catheter is still recommended for patients requiring nutrition for 2 weeks or longer, as well as for those whose fluid restriction requires higher nutrient concentration. We would sooner recommend the new material if catheters, pre-cut at 15 cm, were available to avoid the jagged edges of impromptu modification of commercially available 30-cm ("mid-line") silicone catheters.

Elihu M. Schimmel, MD
William Walker, RN, CNSN
Veterans Affairs Medical CenterBoston, Massachusetts, USA

Elihu M. Schimmel, MD
Veterans Affairs Medical Center
Boston, Massachusetts, USA

William Walker, RN, CNSN
Veterans Affairs Medical Center
Boston, Massachusetts, USA


1. diCostanzo J, Sastre B, Choux R, Kasparian M. Mechanism of thrombogenesis during total parenteral nutrition: role of catheter composition. JPEN. 1988;12:190-4.