© 2003 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume IIA: Biocompatibility, Landes Bioscience, Georgetown, TX, 2003

15.4.2 Geometrical Trapping of Bloodborne Medical Nanorobots

The fate or “clearance” of small immotile particles injected intravenously (IV) into the human bloodstream has been widely investigated. IV injection of, say, 15-micron radiolabeled microspheres is a standard blood flow measurement technique in animal research. The number of spheres that become trapped in a histological tissue slice is proportional to the blood flow through that tissue [2674-2676]. However, particles larger than most capillaries (i.e., 10-, 15-, 25- and 35-micron diameter microspheres) injected into pigs have revealed that there are a considerable number of arteriovenous anastomoses present in the ears and skin, large enough to allow microspheres up to 25 microns to bypass the local capillary bed [2677]. The IV injection of a small number (2-3 million) of 15- to 20-micron microspheres for purely diagnostic purposes in humans is also considered a clinically safe procedure [2678]. (The acute toxicity of 3-micron latex microspheres has been measured in rats [2679]; Section 15.6.2.)

Human capillary vessels average 8 microns in diameter but may be as large as 15-20 microns and as narrow as 4 microns in diameter (Section 8.2.1.2). A rigid particle cannot easily traverse a vessel narrower than the particle’s diameter. Experiments confirm that particles resembling inert nanorobots larger than ~7 microns in diameter are trapped by purely geometrical filtration in the capillary beds [2764]. This trapping usually occurs the first time the microparticles pass through any capillary bed in the body. For example, in one study [2680], 97% of all 15-micron radiolabeled microspheres reaching the canine eye were trapped during the first pass. The trapping of natural red blood cells in capillaries was discussed in Section 8.2.1.2, and the vascular trapping of natural white blood cells [5415-5418] with the possibility of leukergy [4142] or leukoembolization [4143] has also been described in the literature. (There are a few decades-old reports [5419-5421] of leukoembolization in the retina where capillaries are the smallest, and a handful of other reports or suggestions of possible leukoembolization [3890, 5422-5424], but these appear to have minimal clinical significance [4143].) Of course, still-functioning nanorobots can use active motility mechanisms (e.g., microbivores [2762], and see Section 9.4.3) to locomote through narrow vascular passages; localized emissions of vasodilator substances such as NO (nitric oxide) [5884, 5885] could also facilitate such journeys despite efficient scavenging [5886-5888], though NO [5889] and related substances [5890] have many complicating effects which must be carefully considered before they are employed in these circumstances.

At the lower end of the size scale, particles less than 0.1 micron in diameter have the possibility of slipping out of the systemic circulation through fenestrations in the cells lining the blood vessels [2764, 2833]. The fenestrations differ in size for the specific capillary beds present in each organ. For example, the capillary endothelium of pancreas, intestines, and kidney [2681] has fenestrations of 50-70 nm. Exocrine glands also have endothelial walls with 60 nm fenestrations, though these are normally covered by a thin membrane and the basement membrane still presents an intact barrier [2682]. The endothelium of the liver, spleen, and bone marrow has fenestrations of ~100 nm, and the underlying basement membrane is not intact, allowing particles of this size or smaller to escape the vessel lumen. Capillaries in tumor regions may have abnormally high permeability due to tissue inflammation [2764].

Our examination (below) of specific locations where circulating nanodevices might possibly become geometrically trapped suggests that medical nanorobots in the 0.2- to 2-micron size range should have little problem remaining in circulation, if only geometric factors are considered.

Last updated on 30 April 2004