The review, by Shubham Gupta and Varatharajan Prasannavenkadesan of Queen Mary University of London, synthesises studies published between January 2000 and July 2026, drawn from PubMed, Scopus and Web of Science and selected for mechanical relevance. Rather than reviewing suture materials or clinical outcomes in isolation, the authors organise the literature around three tribological interfaces: needle against tissue, suture against tissue, and suture against suture. Their central claim is that low resistance is desirable during penetration and pull-through, controlled sliding is needed while the knot is advanced, and substantial strand-to-strand friction becomes essential once the knot is tightened and must resist slippage.

The needle stage begins before the tissue surface is even broken. Initial contact produces local indentation and viscoelastic deformation, so the reaction force climbs until a critical puncture point is reached. Experimental work on sheep skin and tendon has shown a characteristic force build-up, with skin producing two peaks corresponding to entry through the upper surface and emergence from the underside. Increasing tissue pre-tension raises penetration resistance, while the effect of insertion speed depends strongly on tissue type: varying speed from 1 to 10 millimetres per second had no significant effect in skin, yet faster insertion reduced the force and time needed to puncture liver. Once the needle has passed through, the measured load reflects several mechanisms at once, including cutting at the tip, friction along the embedded shaft and tissue compression generated when the needle deviates from its natural curvature. The review is careful to distinguish total insertion force from friction force proper, noting that the two are often conflated in the literature.

Geometry and trajectory matter enormously. Larger needles and triangular cutting tips produce greater puncture forces than smaller, round-bodied needles, and perpendicular entry gives the lowest resistance among tested angles. Skin and muscle resist insertion more strongly than fat, and vascular structures introduce marked force fluctuations. Perhaps most strikingly, experiments with passive curved needles showed that following the needle’s arc during insertion, rather than advancing linearly, reduced average axial force by 45.87 percent, transverse force by 55.76 percent and bending moment by 57.36 percent in tissue-mimicking gel. Robotic trajectory planning that follows the needle arc and minimises unnecessary tissue displacement, combined with state estimation of deformation and friction forces during passage, offers a route to controlling these interactions in real time.

Once the needle has created a passage, the thread slides against a newly damaged, deformable surface, and the resulting drag includes adhesion, viscoelastic deformation around the suture and, for rough multifilament threads, local ploughing and abrasion. The experimental evidence overturns a common intuition that smoother always means lower friction. In capstan tests, nylon monofilament with a surface roughness of about 4.25 micrometres produced a higher apparent coefficient of friction than coated Vicryl despite being smoother, because smoother monofilaments increase the real contact area with flat substrates and thus adhesive friction. Yet in pull-through testing through a skin substitute, nylon produced a mean drag force of 0.25 newtons against 0.42 newtons for polyglycolic acid multifilament of comparable size, a reduction of roughly 40 percent. Multifilament architecture becomes particularly punishing when individual strands can engage, compress or scrape the walls of the tissue channel.

Diameter, hydration and operating conditions further modulate drag. When polyglycolic acid multifilament diameter increased from 0.19 to 0.55 millimetres, pull-through force rose from roughly 0.35 to 1.15 newtons, although friction did not increase proportionally with the estimated normal force. Increasing normal load from 0.1 to 0.5 newtons reduced the coefficient of friction by 36 percent in one suture-artificial skin pair, while raising sliding velocity from 5 to 25 millimetres per second increased it by 26 percent. Hydration effects are contact-dependent and sometimes counterintuitive: blood and interstitial fluid in porcine tissues reduced adhesion relative to dry artificial skin, whereas fetal bovine serum increased the apparent coefficient for Vicryl from 0.40 to 0.77, raising friction by 88.9 percent for that material while simultaneously damping stick-slip fluctuations. Damage patterns follow the same logic, with polyglycolic acid producing a larger damaged area at the entry boundary than at the exit and retaining more tissue debris than nylon.

Surface engineering offers ways to tune these interfaces, but no single strategy wins everywhere. A plasma-polymerised aminosilane coating applied to curved suture needles lowered penetration force by up to 35 percent compared with conventional silicone-coated controls and remained stable during repeated insertions. A polydopamine-PTFE-activated-carbon coating reduced maximum insertion and extraction forces by 62 and 64 percent in a PVC phantom and by 49 and 30 percent in bovine kidney, with atomic-force microscopy indicating a 65.7 percent reduction in surface friction. For threads, chitosan coating reduced the dynamic coefficient of friction of braided silk while modestly improving tenacity and knot strength. By contrast, dopamine and graphene oxide coatings on polyglycolic acid sutures changed friction very little, with coefficients of 1.37 and 1.42 against 1.38 untreated, even as they improved coating uniformity, wettability and mechanical strength. Antimicrobial treatments can also shift handling: antibacterial processing of braided silk changed suture-suture friction by up to 32.8 percent and increased tissue-drag resistance and bending stiffness by roughly 50 percent.

The knot stage inverts the design goal entirely. A surgical knot does not become secure merely because strands touch; tightening generates normal pressure where filaments cross, allowing suture-suture friction to resist sliding. In polypropylene monofilaments, a dynamic friction coefficient of 0.20 was measured, and knot strength followed a power-law dependence on pre-tension with an exponent of about 1.56. Recent work on general filament mechanics shows that contact pressure within tight knots is strongly localised along curved contact regions, that tension drops sharply at self-contact or nip points where one filament pinches another, and that increased sliding strength with tying tension arises mainly from the simultaneous growth of normal contact force and contact length rather than from plastic deformation. Clinical data reinforce the material dependence: No. 5 FiberWire had the highest failure load at 625 newtons but also the greatest slippage at 14 millimetres, while No. 2 Ethibond slipped only 0.1 millimetres despite lower tensile capacity, and monofilament knots generally require at least five throws with blocking half-hitches.

The review also maps how these interactions are measured and modelled. Capstan and linear-sliding tests provide controlled comparisons of suture surfaces, while the penetration friction apparatus measures needle puncture and suture pull-through sequentially as the skin substitute deforms around the moving thread. Force platforms reveal that novices generate maximum forces of about 4.7 newtons in laparoscopic suturing against 2.6 newtons for experienced surgeons, and that technique matters: closing a standardised 5-newton wound required mean maximum pulling forces ranging from 3.46 newtons for a pulley suture to 8.11 newtons for a horizontal mattress. Finite-element models, from multiscale tendon simulations to spring-based representations of individual sutures on fixed grids, expose contact stresses that are difficult to measure directly. Robotic force feedback has already shown measurable benefits, reducing median maximum force from 3.4 to 2.8 newtons without degrading suturing scores, and lowering mean applied force from 1.53 to 1.04 newtons in robotic interrupted suturing.

The authors’ conclusions point toward a discipline-wide shift. No clinically validated optimal friction range currently exists, because reported values depend on material, surface structure, hydration, tightening force and knot configuration and are not expressed in comparable quantities. The priorities they identify are concrete: standardised wet-tissue testing protocols that specify hydration, pre-tension, insertion angle, pulling velocity and repeated passages; clearer experimental separation of cutting, deformation and friction components; repeated-passage evaluation of coating durability and possible material transfer to tissue; tissue-specific needle and suture selection informed by widely varying damage thresholds, which range from about 1.25 newtons in fallopian tube to 11.43 newtons in fascia; and adaptive robotic control that estimates friction, cutting and deformation separately and adjusts trajectory, speed and grip during each stitch. Friction, in this view, is not an enemy to be eliminated but a resource to be scheduled, spent sparingly during passage and banked heavily at the knot.

Subject of Research: Tribology and interfacial mechanics of surgical suturing, including needle-tissue penetration, suture drag, coatings and knot security

Article Title: Friction and interfacial mechanics of surgical suturing: needle–tissue penetration, suture drag and friction-control strategies

Article References: Friction and interfacial mechanics of surgical suturing: needle–tissue penetration, suture drag and friction-control strategies. (n.d.). https://doi.org/10.3389/fmech.2026.1963772

Image Credits: AI Generated

DOI: 10.3389/fmech.2026.1963772

Keywords: surgical suturing, biotribology, needle-tissue interaction, suture drag, knot security, friction control, surface coatings, medical robotics, finite element modelling, wound closure, suture materials, force feedback