What Medications Need Low Sorbing Tubing?

Several categories of intravenous medications lose a clinically meaningful fraction of their dose to standard plastic tubing, and those drugs should be delivered through low-sorbing alternatives. Nitroglycerin is the most widely recognized example, but the list extends to certain sedatives, insulin, immunosuppressants, some chemotherapy agents, and select protein-based therapies. The problem is not just theoretical: in the first minutes of an infusion through conventional polyvinyl chloride (PVC) tubing, a patient receiving nitroglycerin may get barely more than half the intended dose. The type of plastic the tubing is made from, how fast the fluid flows, and how long the tubing is all change how much drug actually reaches the bloodstream.

Nitroglycerin and the Drug That Started the Conversation

Nitroglycerin is the textbook case for tubing sorption problems. When infused through standard PVC tubing containing the plasticizer DEHP (di-2-ethylhexyl phthalate), roughly 57% of the initial nitroglycerin concentration reaches the patient in the opening minutes of the infusion. That means nearly half the drug is absorbed into the tubing wall before it ever enters a vein.¹ The drug gradually saturates the plastic, so concentrations creep upward over time, but they never become truly steady. That creates an unpredictable dose-response relationship, which is a serious problem for a medication used to control blood pressure and chest pain in acute settings.

A pharmacokinetic comparison across tubing types found that when nitroglycerin was delivered through polyolefin (PO) tubing, patients achieved substantially higher plasma drug levels than when the same dose was run through PVC or polyurethane (PU) sets. The difference was large enough to show up in blood pressure readings: patients on PO tubing had measurably greater drops in mean arterial pressure than those on PU tubing at most time points during the infusion.² In practical terms, this means a clinician titrating nitroglycerin through PVC tubing is fighting both the patient’s physiology and the tubing’s appetite for the drug. Switching to polyolefin or glass-lined delivery removes one of those variables.

Isosorbide dinitrate, another nitrate vasodilator in the same pharmacological family, follows a similar sorption pattern to nitroglycerin when modeled against PVC tubing.³ Any nitrate given intravenously should be treated as a candidate for low-sorbing tubing.

Diazepam and Related Sedatives

Diazepam is among the worst offenders for sorption loss. In static contact experiments, PVC tubing absorbed so much diazepam that after 96 hours of contact, only about 1.3% of the original drug concentration remained in solution. Even under dynamic flow conditions meant to simulate real clinical use, PVC and PVC/PU tubing cut the diazepam concentration to roughly 15% of its starting value within the first measurement.⁴ That is a staggering amount of drug simply disappearing into plastic.

The tubing material makes a dramatic difference. Polyolefin-based tubes delivered more than 90% of the administered diazepam, while PVC and polyurethane tubes showed far higher sorption.⁵ Tubing length also matters: in one study, the concentration of diazepam exiting the tube ranged from about 45.5 micrograms per milliliter using a short 23-centimeter PVC segment down to 28.8 micrograms per milliliter using the full 185-centimeter set, despite starting from the same solution.⁶ The percentage of drug adsorbed also increased at shorter infusion durations, because the plastic had not yet approached saturation.

Chlorpromazine, an antipsychotic sometimes given intravenously, follows a sorption pattern that requires a more complex diffusion model than nitroglycerin, involving both diffusion into the plastic and an interfacial resistance at the liquid-plastic boundary.³ The practical takeaway is the same: standard PVC tubing absorbs it. Interestingly, some drugs that might seem vulnerable are not. Lorazepam, midazolam, promethazine, and thiopental all remained stable in PVC containers in at least one study, showing that sorption risk cannot be guessed from drug class alone.⁷

Insulin

Insulin is unusual in this group because the dominant concern is surface adsorption rather than absorption into the plastic matrix. The protein sticks to surfaces it contacts, and this behavior depends heavily on concentration and flow rate. At low concentrations and slow flow rates, insulin loss to the tubing wall can be meaningful. Research has shown that insulin loss was less at a flow rate of 10 mL per hour compared to 1 mL per hour, likely because faster flow saturates the binding sites on the tubing surface more quickly, after which concentrations stabilize.⁴

One study specifically examining dwell time in polypropylene IV sets found that how long the insulin solution sat in the tubing before flowing did not significantly affect adsorption at any of the three concentrations tested.⁸ This suggests that for polypropylene specifically, stagnant fluid in the line is less of a concern for insulin than it might be for lipophilic small molecules. Still, the overall behavior of insulin at the tubing wall varies enough with concentration and tubing material that many hospitals use low-sorbing or dedicated insulin infusion sets as standard practice.

Unlike nitroglycerin and diazepam, where PVC is consistently the worst performer, insulin behaved somewhat differently across materials: PVC tubing actually induced the least insulin loss among all tested materials in one comparative study.⁴ This counterintuitive finding highlights that “low-sorbing” is not a universal label. What works for one drug may not work for another, and the recommendation for a particular tubing material has to be drug-specific.

Tacrolimus and Immunosuppressants

Tacrolimus, widely used to prevent organ rejection after transplant, is another drug with documented sorption to IV tubing. It has been used alongside diazepam as a model compound in sorption studies precisely because it is known to stick to PVC and other polymeric surfaces.⁹ For transplant patients, where maintaining precise blood levels of immunosuppressant drugs can mean the difference between organ acceptance and rejection, even modest losses to tubing are unacceptable. Cyclosporine, another calcineurin inhibitor used in transplant medicine, is similarly flagged as a drug that interacts with IV delivery materials.

Paclitaxel and the Plasticizer Leaching Problem

Paclitaxel (marketed originally as Taxol) presents a different but related concern. Its formulation vehicle contains Cremophor EL and ethanol, both of which aggressively leach DEHP from PVC bags and administration sets.¹⁰ This is not sorption of the drug into the plastic but rather extraction of the plasticizer out of the plastic and into the drug solution, which the patient then receives intravenously. DEHP is classified as an endocrine disruptor and a reproductive toxicant, so exposing patients to it is a clear safety concern.

Testing has found that the paclitaxel vehicle leached DEHP not only from PVC sets but even from some administration sets labeled as non-PVC.¹¹ This is an important caution: a “non-PVC” label does not guarantee the absence of DEHP-containing components in connectors or other parts of the set. Paclitaxel must be administered through DEHP-free tubing and bags, and the institution should verify the entire delivery path, not just the main line.

Amiodarone

Amiodarone, a potent antiarrhythmic used for life-threatening heart rhythm disturbances, has historically come with manufacturer instructions to use glass bottles and non-PVC lines. The reality turns out to be more nuanced than the label suggests. One study comparing PVC bags to glass bottles found only about a 2.7% loss of amiodarone in 500 mL PVC bags over 24 hours, and no detectable loss in 100 mL PVC bags. When PVC administration sets were attached to glass bottles, the maximum adsorption loss in the outflow was around 4.9% within the first hour.¹² The researchers concluded these results did not support the blanket recommendation to avoid PVC for amiodarone. A separate study found an average recovery of about 91.5% of amiodarone after delivery through an IV administration set.¹³

So amiodarone does lose some drug to plastic, but the magnitude is small enough that some researchers have questioned whether the expensive precaution of glass-only delivery is justified. Many hospitals still default to non-PVC tubing for amiodarone out of an abundance of caution and because the drug label says to, but the evidence suggests the clinical significance of the loss may be modest compared to drugs like nitroglycerin or diazepam.

Protein-Based Therapies and Monoclonal Antibodies

Biologic drugs, including monoclonal antibodies and fusion proteins, face a related but mechanistically distinct challenge. Rather than absorbing into the plastic bulk, these large molecules tend to adsorb onto surfaces they contact, including tubing walls and especially in-line filters. Research examining four different therapeutic proteins (two IgG1 antibodies, one IgG4, and one Fc fusion protein) found that adsorption depended more on the dilution solution and filter material than on the proteins’ own properties. Protein loss was higher when filters had a strongly negatively charged surface, particularly when the drugs were diluted in glucose solution rather than saline.¹⁴

For biologics, the concern is often the filter rather than the tubing itself. Many biologics are administered at very low concentrations, where even a small absolute amount of protein sticking to surfaces represents a large fraction of the dose. Manufacturers of individual monoclonal antibodies typically specify compatible administration sets in their prescribing information, and those recommendations should be followed closely.

Why Some Drugs Stick and Others Do Not

The physical chemistry behind sorption helps explain the pattern. Drugs that are small, lipophilic (fat-soluble), and electrically neutral tend to partition readily into the soft, plasticized matrix of PVC tubing. Nitroglycerin and diazepam both fit this profile. The plasticizer DEHP, which makes PVC flexible, actually creates a more hospitable environment for these molecules by increasing the free volume within the polymer. Research into polyurethane catheters has confirmed that the internal structure of the plastic matters: polyurethanes with more phase-mixing between their hard and soft segments had less free volume, which slowed drug diffusion into the material.¹⁵ The dominant force driving drug-polymer adhesion was short-range molecular attraction rather than any chemical bonding.

Water-soluble, charged, or large molecules tend to stay in solution rather than entering the plastic. That is why many common IV antibiotics, electrolytes, and hydrophilic drugs are not flagged for sorption concerns. But the transition is not sharp: some moderately lipophilic drugs show intermediate behavior that depends on concentration, contact time, and the specific polymer blend.

How Flow Rate and Tubing Length Change Drug Delivery

Even when a drug is given through the right tubing material, operational details alter how much drug reaches the patient. Slower flow rates give the solution more contact time with the tubing wall per unit volume, increasing the fraction of drug absorbed. For both paracetamol and diazepam, drug loss was greater under static (zero flow) conditions than at either 1 mL/h or 10 mL/h, and the 10 mL/h rate outperformed 1 mL/h for insulin delivery as well.⁴ In practice, this means that neonatal or pediatric infusions, which often run at very low flow rates through relatively long tubing, face a disproportionate sorption risk.

Tubing length has a straightforward effect: more plastic surface means more drug loss. The diazepam study comparing tubing lengths found a nearly 40% difference in drug concentration between the shortest and longest segments at the same flow rate.⁶ Keeping tubing as short as practically possible is a simple intervention that reduces drug loss regardless of material.

Flushing the line after an infusion is another often-overlooked variable. Drug residue left in the administration set after the infusion bag empties represents an undelivered portion of the dose. For antimicrobials, failing to flush can mean the patient does not reach the blood levels needed to treat the infection, potentially prolonging illness or contributing to resistance.¹⁶

What “Low-Sorbing” Tubing Actually Means

The term “low-sorbing” is not a regulated standard with a single definition. In general, it refers to tubing made from materials that absorb significantly less drug than traditional plasticized PVC. The most common alternatives include polyolefin (PO), polyethylene-lined PVC, and certain thermoplastic elastomers like SEBS (styrene-ethylene-butylene-styrene). These materials lack the DEHP plasticizer that softens standard PVC and simultaneously creates a reservoir for lipophilic drugs.

Polyolefin has emerged as the best-performing material for delivering drugs like diazepam and nitroglycerin with minimal loss.⁵ SEBS tubing also performed well for diazepam in comparative testing.⁴ Polyurethane, however, is an unreliable alternative: it outperformed PVC for some drugs but actually produced worse sorption for diazepam and dramatically lower nitroglycerin delivery than polyolefin.² Buying “non-PVC” tubing without checking which non-PVC polymer it actually uses is not enough.

The healthcare industry has been gradually moving away from DEHP-plasticized PVC for IV delivery in general, driven partly by sorption concerns and partly by toxicological worry about DEHP itself. A study assessing critical care nurses’ knowledge found that the shift toward DEHP-free lines was already reshaping how “standard” versus “special” lines were being defined in hospitals.¹⁷ In many facilities, what used to be the “special” low-sorbing line is becoming the new default.

When Nurses and Pharmacists Get It Wrong

Knowledge gaps around tubing compatibility remain a real clinical problem. The same nursing assessment study revealed that many critical care nurses did not correctly identify which medications required specific line types.¹⁷ This is understandable: the drug list is not always intuitive (lorazepam is fine in PVC; diazepam is not), the tubing options keep expanding, and product labeling varies between manufacturers and countries.

The consequences of getting it wrong depend on the drug. For a nitroglycerin drip in a cardiac patient, under-delivery can mean inadequate blood pressure control during a heart attack. For diazepam used to stop seizures, losing the majority of the dose to tubing can mean the seizure continues. For an immunosuppressant like tacrolimus in a transplant recipient, subtherapeutic levels can trigger organ rejection. The stakes are highest when the drug has a narrow therapeutic window, when the patient is critically ill, and when the infusion rate is slow enough to maximize contact time with the tubing.

PVC has persisted as the dominant IV tubing material for decades because it is cheap, flexible, transparent, and easy to manufacture. The shift to alternatives involves higher material costs and sometimes different handling characteristics. But as the evidence base for sorption-related dose loss has grown, and as concerns about DEHP exposure have added a second reason to move away from plasticized PVC, the economic calculation is changing. Hospitals increasingly weigh the cost of a slightly more expensive tubing set against the cost of an extended ICU stay caused by a drug that did not reach its target concentration.

A Quick Reference of High-Risk Medications

While this list is not exhaustive and institutional formularies vary, the following drugs are most commonly identified as needing low-sorbing or DEHP-free delivery systems:

Nitroglycerin: large and unpredictable sorption to PVC; polyolefin strongly preferred.
Isosorbide dinitrate: similar sorption profile to nitroglycerin in PVC tubing.
Diazepam: extreme sorption to PVC and polyurethane; polyolefin delivers over 90%.
Insulin: adsorption to surfaces varies by concentration and flow rate; low-sorbing sets commonly used, especially for low-dose infusions.
Tacrolimus: documented sorption to PVC-based tubing; non-PVC delivery recommended.
Cyclosporine: interaction with IV delivery materials flagged in drug labeling.
Paclitaxel: vehicle leaches DEHP from PVC; must use DEHP-free tubing and bags.
Chlorpromazine: sorption to PVC follows a complex diffusion-plus-resistance pattern.
Amiodarone: modest losses documented; many hospitals still use non-PVC per label, though evidence suggests clinical impact may be small.

For any IV medication you are unfamiliar with, the most reliable starting point is the drug’s own prescribing information, which will specify compatible containers and administration sets if sorption or leaching is a known concern. When in doubt, pharmacists are typically the best resource in a hospital setting for matching a drug to the right tubing. The broader trend in the field is toward replacing plasticized PVC as the default for all infusions, which would eventually make the question of which drugs need special tubing moot.

1
PubMed. Investigation into the sorption of nitroglycerin and diazepam into PVC tubes and alternative tube materials during application
2
PubMed Central. A comparison of the pharmacokinetic and pharmacodynamic properties of nitroglycerin according to the composition of the administration set: A preliminary study
3
PubMed. Modeling solute sorption into plastic tubing during organ perfusion and intravenous infusions
4
Scientific Reports. Impact of alternative materials to plasticized PVC infusion tubings on drug sorption and plasticizer release
5
PubMed. Diazepam sorption to PVC- and non-PVC-based tubes in administration sets with quantitative determination using a high-performance liquid chromatographic method
6
American Journal of Health-System Pharmacy. Effect of tubing length on adsorption of diazepam to polyvinyl chloride administration sets
7
American Journal of Hospital Pharmacy. Sorption of various drugs in polyvinyl chloride, glass, and polyethylene-lined infusion containers
8
PubMed Central. The effect of tubing dwell time on insulin adsorption during intravenous insulin infusions
9
PubMed Central. Evaluation of Drug Sorption to PVC- and Non-PVC-based Tubes in Administration Sets Using a Pump
10
International Journal of Pharmaceutics. The extraction of diethylhexylphthalate (DEHP) from polyvinyl chloride components of intravenous infusion containers and administration sets by paclitaxel injection
11
American Journal of Hospital Pharmacy. Compatibility of paclitaxel injection vehicle with intravenous administration and extension sets
12
PubMed. A comparative analysis of the loss of amiodarone from small and large volume PVC and non-PVC infusion systems
13
PubMed. Adsorption and Leachable Contamination of Flucloxacillin, Cyclosporin and Amiodarone Following Delivery Through an Intravenous Administration Set
14
PubMed. Protein Adsorption to In-Line Filters of Intravenous Administration Sets
15
PubMed. Microphase Structure and Component Influence on Drug Sorption: Polyurethane-Drug Interaction Elucidation in Medical Catheter Systems
16
PubMed Central. Understanding IV antimicrobial drug losses: the importance of flushing infusion administration sets
17
PubMed Central. Critical Care Nurses’ Knowledge of Correct Line Types for Administration of Common Intravenous Medications: Assessment and Intervention Study