Your casualty is elderly. Somebody finds a pill organizer, or a medical alert bracelet, or a family member says the word apixaban. And a piece of training surfaces in your head: kaolin needs the clotting cascade, this drug has switched part of that cascade off, so kaolin will not work here. Reach for the chitosan.

We have taught that. Plenty of good instructors still do. It is a clean argument, it flows from real physiology, and it is the kind of reasoning that feels like competence.

It is also not supported by the evidence, and it is not in any guideline. This post walks through what is actually known, what is not, and what to do with your hands while you are standing over the patient.

First, give the other side its best argument

This teaching did not appear from nowhere, and it deserves a fair hearing before we take it apart.

Kaolin works by contact activation. It is an aluminum silicate that initiates the intrinsic coagulation cascade at factor XII. Chitosan works differently — it is a cationic polysaccharide that drives red cells and platelets to aggregate and adheres to wet tissue, through a mechanism that is largely independent of the cascade entirely.

So if a drug inhibits factor Xa, an agent that acts upstream of factor Xa should have very little to offer. That prediction is coherent enough that a research team formally wrote it down in advance as their study hypothesis. And when chitosan was added to the TCCC guidelines in the 2014 change document, the stated rationale did cite its performance in coagulopathic conditions.

None of that is marketing. It is real physiology and a real guideline document. The error came later, when instructors converted it into a selection rule the guideline never wrote.

What happened when somebody measured it

The study that matters most

A prospective randomized trial enrolled twenty-six patients who continued their anticoagulation straight through a minor debridement procedure. They were randomized to kaolin-impregnated gauze or to standard sterile dry gauze.

At five minutes, the kaolin group had achieved hemostasis in eighty percent of cases, against forty percent for plain gauze. At ten minutes, the kaolin group was at one hundred percent. Plain gauze was at fifty-eight point three percent. The difference was statistically significant, and no adverse effects were reported.

Pay attention to the comparator, because it is what makes this study load-bearing. It was plain gauze — which means the study isolated what the kaolin itself contributes, separately from what the textile contributes, in living patients whose blood was pharmacologically thinned.

The study that predicted failure and did not find it

Eight emergency department patients taking rivaroxaban had blood tested by viscoelastic assay against kaolin and chitosan agents. As noted above, the investigators pre-registered the hypothesis that kaolin would fail, on exactly the mechanistic grounds described earlier. Both agents trended toward improved coagulability. The kaolin-fails hypothesis was not supported.

Eight patients is a small number and a trend is not a result. But a pre-registered prediction that fails is worth more than an unplanned finding that succeeds, and this one failed cleanly.

The inert gauze that scored at the top

Here is the part that should make you cautious about this entire category of testing, including the studies that support our own position.

A third study ran a full shelf of hemostatic gauzes against blood drawn from patients on enoxaparin, on heparin with aspirin, on apixaban, and on phenprocoumon. Every agent tested improved the onset of clotting. The authors concluded that the idea kaolin is ineffective in anticoagulated blood is inaccurate.

That sounds like a strong result for our argument. Then look at what else was in the sample set.

One of the tested items was the manufacturer's own training gauze — the inert practice version. Not for medical use. No kaolin. No hemostatic agent of any kind. It is a plain white gauze strip built to feel like the real product for repetition drills.

That training gauze placed in the top performance tier — alongside the live kaolin product, and ahead of the chitosans.

An inert textile cannot outperform a chitosan by pharmacology. So a substantial part of what that assay detected was not chemistry at all. It was the effect of introducing a large-surface-area foreign material into a test cup that runs on contact activation.

What the guideline actually says

This is worth stating precisely, because the misquotation is common and it has ended up printed on commercial product pages.

The TCCC Guidelines of 25 January 2024 name Combat Gauze as the CoTCCC-recommended hemostatic dressing of choice. Celox Gauze, ChitoGauze, XStat and iTClamp are listed as alternative hemostatic adjuncts. XStat is specified for deep, narrow-tract junctional wounds. At least three minutes of direct pressure is required, with XStat excepted.

There is no anticoagulant-based selection rule anywhere in that guideline. None. No drug-history branch, no coagulopathy branch, no instruction to swap chemistries based on a medication list. Any content telling you CoTCCC prefers chitosan for anticoagulated patients is inventing a guideline position.

What the guideline does specify is an escalation rule — and notice that it is failure-driven, not drug-driven. If a dressing fails, you remove it and repack with a fresh dressing of the same type or a different type. The trigger is the bleeding continuing, not the pharmacy record.

What is still unknown

We hold ourselves to naming the gaps, so here they are.

One more distinction that gets blurred constantly and should not be. Pharmacologic anticoagulation and trauma-induced coagulopathy are not the same problem. A patient on apixaban and a hypothermic, acidotic, hemodiluted trauma patient have different lesions, and evidence from one does not automatically transfer to the other. A great deal of the swine coagulopathy literature uses dilutional and hypothermic models. That work is valuable, and it is not evidence about a patient on a factor Xa inhibitor.

What to do at the wound

Strip all of the above away and you are left with a short list, in order.

1. Expose the wound. You cannot pack what you cannot see, and the single most common failure in wound packing is packing into the general vicinity of the injury rather than onto it.
2. Pack directly onto the bleeding source. Find the vessel with your fingers. Drive the gauze against it.
3. Use enough material. Fill the tract. Under-packing is the second most common failure, and it does not get fixed by a better molecule.
4. Hold sustained direct pressure for a full three minutes. Not a hopeful ninety seconds while you look around for the next thing.
5. Reassess, and expect trouble. Anticoagulated casualties re-bleed more often, regardless of what you packed with. Check it again. Then check it again after you move the patient.
6. If it fails, repack. Pull the dressing, pack fresh, same type or different. Escalate mechanical control. If systemic anticoagulant reversal is available downstream, that becomes part of the plan — but it is a hospital intervention, not a reason to slow down in the field.

Nowhere in that list does anyone stop to determine which anticoagulant the patient is taking. That information matters enormously to the receiving facility. It does not change what your hands should be doing in the next four minutes.

Sources

• Hwang et al. Effectiveness of kaolin-impregnated hemostatic gauze in anticoagulated patients undergoing minor debridement: a prospective randomized trial. J Diabetes Res. 2019;2019:9316380. PMID 31828170. CRIS KCT0003894.
• Bar J, David A, Khader T, Mulcare M, Tedeschi C. The effect of hemostatic agents on coagulation in patients taking rivaroxaban. Prehosp Disaster Med. 2017;32(5):580–587. PMID 28625215.
• Lechner R, Hanke K, Schmid A, Mayer B, Helm M, Kulla M, Hossfeld B. Hemostatics in patients with inhibited coagulation (HEMPIC). Transfusion. 2023;63(Suppl 3):S159–S167. doi:10.1111/trf.17333. Cited here with the inert-trainer caveat described above.
• Bennett BL, Littlejohn LF, Kheirabadi BS, Butler FK, Kotwal RS, Dubick MA, Bailey JA. TCCC Guidelines-Change 13-05. J Spec Oper Med. 2014. PMID 25344707.
• Littlejohn LF, Devlin JJ, Kircher SS, Lueken R, Melia MR, Johnson AS. Acad Emerg Med. 2011;18:340–350.
• Kozen BG, Kircher SJ, Henao J, Godinez FS, Johnson AS. An alternative hemostatic dressing: comparison of CELOX, HemCon, and QuikClot. Acad Emerg Med. 2008;15:74–81.
• Systematic review with meta-analysis: chitosan hemostasis in dental patients on antiplatelet/anticoagulant therapy. PMC10782677.
• TCCC Guidelines, 25 January 2024. Joint Trauma System.

Educational content. Does not replace the current TCCC Guidelines, your local protocol, or your medical director's direction. Where the evidence is thin or contested, we said so rather than rounding it off.