Field of the Invention

The present invention concerns a collagen or catgut suture fila¬ ment coated with a protective flexible sheath made of an adherent polymeric resin that can be degraded hydrolytically but is enzymatic- ally stable. It also concerns a method for its preparation.

Back-ground of the Art.

It is well kncwn that ordinary catgut is currently losing atten- tion as a surgical suture filament because of its biodegradation pro¬ perties which are not favorable. Indeed, it has been shcwn that, when catgut is in contact with the living tissues surrounding a wound which has been stitched, it degrades, enzymatically and loses its mechanic- al properties guite rapidly. This drawback can be sαnewhat lessened by subjecting the oollagen of the catgut to tanning with chromium salts, hσwever, such procedure has disadvantages since chromium com- pounds are toxic. Further, catgut sutures, chromated or not, induce unwanted tissue reactions, especially for the first days, such react- ions being much less significant with more modern suture materials.

Thus, there is now a tendency to more and more replace the cat ^¬ gut sutures by synthetic filaments or braids made of polyesters the degradation of which has a profile different fron that of catgut since it is not enzyme catalyzed. In short, when such polyesters (polyoxy- acetyl esters such as polyglycolic or polylactic esters) are used as suture material, they will retain their tensile properties long¬ er than catgut (even chrcmium treated) although the overall resorption time is about the same. Details on these questions will be found in the follσwing references: ENCYCLOPEDIA OF POLYMER SCIENCE & TECHNOLO- GY, Vol 1 (Supplement) , p. 587-596. P.Y. WA G et al: Structural Re- quirements for the Degradation of Condensation Polymer in Vivo, Poly¬ mer Science & Technology, Plenum Press (1973) . K. SÖGIMACHI et al: Evaluation of Absorbable Suture Materials in Biliary Tract Surgery, CA 89, 30728y. E.L. HOWES: Strength Studies of Polyglycolic Acid

OMPI 'IPO

versus Catgut Sutures of the Same Size, CA 79_- 57648c. A.B. KOVACS et al: Ccmparative Study of Tissue Reactions to Various Suture Ma¬ terials (Catgut, Silk and Polyesters) , CA 72, 11183z.

Objects of the Invention

However, suture filaments made of synthetic polymers are pre- sently much more expensive than catgut made of ruminant guts and it is highly desirable that catgut be revived by simple, effective and cheap means. Such means are an object of the present invention which provides a catgut suture filament coated with an adherent protect- ive sheath made of a resin that is hydrolytically degradable but which is thick enough to shield the σollagen catgut core fron the action of enzymes for a period sufficient for the sutured wound to heal be- fore the suture filament loses much of its intrinsic mechanical pro¬ perties. In practice, the sheath can have a thickness of fron a few tenths of a micron to several hundredths- of microns, however, this thickness can be adapted at will depending on the needs. When a so- re is stitched with such a sheathed suture filament, the sleeve de- grades very slowly such that the wire retains its mechanical proper¬ ties for a time sufficient for the wound to heal; then, when the sheath has finally been hydrolyzed, the catgut core is attacked by the body enzymes and is resorbed very quickly.

Another object of the present . invention is to provide a sutu- re filament with a hcmogeneous and soft surface which is substantial- ly free fron pinholes, such pinholes being possible Sites for the body fluids enzymes to penetrate the sheath and degrade the filament.

Another object of the invention is to provide a sheathed cat¬ gut filament with σontrolable sheath resistance to hydrolysis, such control being possible by properly selecting the polymer resin of such sheath and the σoating oonditions.

Another object of the invention is to provide a suture filament with excellent surface properties such as "slickness" and "knot-pull" properties as well as resistance to "knot slippage" when moistened by body fluids.

Still another object of the invention is to provide suture fila¬ ments with sheaths that will slowly degrade during a first period

OMPI 4_ v.'i?o

of time sufficient for the wound to heal, the mechanical properties of the filament staying substantially constant for such period, and that will then be rapidly resorbed by the organism during a second period after which the suture remnants are naturally eliminated. Another object of the invention is also to provide methods for manufacturing the above mentioned sheathed filaments.

Other objects will beccme apparent fron the following detail- ed descripticn of the invention.

Detailed Description of the Invention.

As a resin suitable for the sheath of the present suture fila¬ ment, various polymers can be considered, the requirements being that they form thin flexible layers substantially resistant to enzyme de- gradation and only slowly hydrolyzed by the body tissues. As such, for instance, polyesters, preferentially reinforced by urethane and urea links, can be used.. The base polyesters can have a general struct- ure similar to that used for known synthetic sutures (polyhydroxy- acids-esters) or be made fron the polycondensation of selected diol- ccmpounds with selected diacid-compounds. Among the diacids, oxalic acid, malonic acid, succinic acid, glutaric acid, adipic acid, pi- elic acid, perfluoroadipic acid, 2,2-oxidiacetic acid, 2-oxogluta- ric acid, D-tartaric acid and the like are σonvenient. Among the diols, the follcwings can be recited: aliphatic glycols of from 4 to 12 carbon atαns such as 1,4-butane-diol, 1,6-hexanediol and poly- alkylene glycols (e.g. polyethylene glycols and polypropylene gly¬ cols) having 2 to 15 polyalkylene glycol units. It is important to note that the properties of the sheath are dependent on the proper selection of the diacid and the diol and also on some degree of cross- -linking which can be introduced, if desired, as described herein- after. In other words, the polymer can be made more or less flexible, more or less resilient, more or less smooth and more or less resist¬ ant to hydrolysis depending . on the diacid selected, on the length of the alkylene or polyoxylalkylene Segments and, of course, on the polymerization stoichic etry (molecular weight of the polyester diol resulting frαti the polycondensation) . As an example of the above des¬ cribed possible variations, it can be mentioned that, when using a

OMPI

given diol, a sheath made from a polyester-polyurethane-polyurea re¬ sin containing oxalic acid is less resistant to hydrolysis than the corresponding resin containing, instead, 2,2-oxydiacetic acid, the latter being itself less resistant than the corresponding polymer based on glutaric acid.

Using a proper molecular ränge for the base polyester is also an i portant factor in the present invention. Being considered that the polycondensation can be schematized as follows:

A1C1 ₃ [n + l] HO - R - CH + [ή] HOOC - R' - COCE — -^ » ^•

HO-^R-CKD-R'-OCH^ROH (I), it results that the base polyester diol molecular weight ( H^) will depend on the ratio n/n + 1 and the nature of R (the alkylene or polyoxyalkylene segment of the diol) and of R' (the linking segment of the diacid) . Thus, for instance, using 0.373 mole of 2,2-oxydiacetic acid and 0.391 mole of diethyl- eneglycol gives, after calculation frαn the above formula n n + 1 = 373/391, n = 20 frσ which the molecular weight M^o; of formula I (R = R" = -(CH ₂ ) ₂ -0-(CH ₂ ) ₂ -; M^ = MW _R , = 80) is found to be 5074. In general, number average molecular weights determinations by anal- ysis are in rather near agreement with the above predicted values.

In the practice of the present invention, high molecular weights

(i.e. n being in the ränge of 15 or more) are not advantageous un- less sαne degree of cross-linking is present. Such cross-linking can be introduced by using, in admixture with the diol, a proportions of a polyol, e.g. a triol, a tetrol or compounds containing more than three hydroxygroups; exa ples of such compounds are glycerol, tri- methylol propane or hydrogenated sugars such as hexoses or pentoses, e.g. sorbitol. Indeed, in the absence of cross-links, polyestergly- σols with molecular weights in the higher ränge (above 15 or more) will provide sheaths with sαnewhat too much elasticity; in other words, they will elongate too much under use stress relative to the catgut collagen core which effect is sometimes inαonvenient. This can be remedied by either keeping n relatively low, e.g. between 2 and 10, preferably, or, when using polyesters glycol with n above 10 or more, adding frαn about 0.5 to 10% of the above triols or poly- ols. Naturally, in the practice of the present invention, mixtures

of two or several diacids and/or diols can be used for preparing the base polyester glycol.

In the above polyester glycol condensation reaction scheme, it has been indicated that the condensation can be effected by heat in the presence of a catalyst (AlCl ) . it should be well understood that the conditions for effecting the polycondensation and obtaining the required polyester glycols are not novel per se and that classical polyester manufacturing conditions can be widely used such as a ränge of usual catalysts (or no catalyst, if suitable) and a ränge of usual polyesterification temperature and reaction times known to ordinary chemists.

The polyesters glycols used in the present invention can also be characterized by the OH nu ber (No _QH ) which amounts to the number of mg of potassium hydroxide corresponding to the quantity of -OH groups in one gram of the polymer. Thus, this value can be obtain- ed frcm the inverse of the molecular weight M ^ times twice the ole-

Url cular weight of KOH times one thousand. ^' This value of No U _™ -ti. can also be obtained by analysis fron a method in Anal. Chem. 35 (4), 571 (1963) which consists in weighing exactly an aliquot of the polyes- ter-diol (m gram) , blocking the CH groups by a kncwn excess of phe- nylisocyanate, destroying the excess isccyanate with a known quan¬ tity of dibutylamine and back-titrating the excess dibutylamine with a HCIO. solution. The analysis is carried out together with a control blank containing no polyester-diol; thus if x defines the a ount of mole of HCIO^ less used in the blank than in the sample (i.e. x = x - x^ ml if HZIO, is normal and the sample has consumed x ml and the blank has oonsumed X _j ^ ml of such HC10 ₄ solution) , then o _0H = 56.1x/m (MWJ _T QJ = 56.1). Thus, the above described measurement of the CH number is a convenient way to measure the molecular weight of the polyester diol for correlating with the molecular weight value derived fron the relative proportions of the glycol and the diacid used in the polycondensation. The relationship is evidently

MW = 56.1 x 2000—

OH No _QH

To ter inate with the polyester-diol preparation, it should still be mentioned that other techniques of preparation have been tested and are possible although less preferred, such as the transesterifi¬ cation ethod in which the starting diol is heated with a lewer es- 5 ter of the diacid and the resulting lower alcohol formed is removed by distillation. This has been tried with diethylene glycol and di- methyl oxalate, the MeQH formed being removed under vaeuum. It should also be mentioned that glycol prepolymers can be found cαnmercial- ly, all ready, and can be used in the present method in place of the

10 polyesters made as described above. Also, other polyesters usable in the present invention are described in detail in IBP 3,778,390. Generally speaking, the ränge of the polyesters glycols preferably usable in the invention have molecular weights between about 250 and 10,000.

15 For binding the above polyesterdiols to the catgut core, i.e. to preduce a coated sheath on the catgut filament that will proper¬ ly adhere thereto, isoeyanates are preferably used to cap the diols since they can fulfill three main funetions: First they can attach to the collagen either by Van der Waals forces or by hydrogen bonds

2.0 (between the urethane or urea groups derived, after reaction, fron the NGO groups and the polypeptide amide funetions of the collagen) or by covalency through the reaction of the NGO groups with the free -NH ₂ groups (lysine) or free OH groups (hydroxyproline) of said colla¬ gen. Thus, in the practice of the invention, the polyester-diol will

25 then be capped with diisoeyanates (or polyisoeyanates if further cross- linking is wanted) .

The second funetion of the isoeyanates is to produce se e de- gree of chain extension provided the quantity of diisoeyanates used is sαnewhat less than twice the stoichiαnetry relative to the diol.

30 A third funetion of the isoeyanate is to enable what is called " oisture curing" to oeeur α the freshly coated filament cn Stand¬ ing. This operates when the coated fil still contains an excess of unreacted NCO groups which can then react with ambiant moisture as follows (R" being any conceptual radical of the isoeyanate cαnpound) : 5

Rü-NOO + H Q →- R" ML, + 00. and

Rϊ-NGO + ^N-R" > R'^-NH-CO-NH-R"

Moisture curing procures a smooth "drying" of the coated finish and contributes also to the excellent properties of the present sheath. In the present invention, a variety of diisocyanates and poly- isocyanates can be used. Aliphatic and cycloaliphatic diisocyanates can be used such as hexane-diisoeyanate or cyclohexane-diisoeyanate; however, arornatic diisocyanates are preferred because of their higher reactivity: Convenient diisocyanates are p-phenylene diisoeyanate, 2,6- and 2,4-toluene diisoeyanate (TDI) or p-tolylene diisoeyanate. Other suitable di- or poly-isoeyanates are disclosed in British Patent No. 1,430,422, page 3.

The amount of isoeyanate cempound to be used relative to the diol prepolymer depends on the needs and on the end properties to be given to the sheath. In general, suitable mole ratio of diisoeya¬ nate to diol is between 1.1 and 1.5. At lσwer ratio, the chain ex- tension may become too great before capping becomes effective and the resulting diioscyanate polymer may becαne too thick for proper coating uses. At the other end, using diisoeyanate/diol ratios high- er than 1.5 may lead to an excess of free isoeyanate groups in the coating material and the end formation of too any urea links after moisture curing which may result in too much rigidity in the sheath. However, the above values are only indicative and may be exceeded, if desired, in both directions. Generally, the reaction of the prepolymer diol and the diiso¬ eyanate is performed by ixing the ingredients at roem temperature in a suitable solvent and in the presence or in the absence of a ca¬ talyst. A catalyst is advantageous if the reaction must be speeded up but it is not strictly indispensable. As catalysts, tin compounds such as tin oetanoate or diaminobicyclcoctane (DABCO) of formula

N — CH ₂ —CH ₂ N

^ CH ₂ —CH ₂ ^

can be used advantageously although other common isoeyanate react¬ ion catalyzing materials can also be used. The amount of catalyst

is in the ränge of 0.1 to 5%.

As solvents, the follcwing can be used: toluene, tetrahydrofu- rane (THF) , dioxane, dimethylformamide (DMF) , diglyme, ethyl acetate, acetone, cellosolve acetate, methyl-ethyl ketone, pyridine, etc. The solvents must be chosen depending on the needs and on the pre¬ polymer properties, keeping in mind that the higher the molecular weight of the polyester glycol, the less soluble it becαnes. Methyl ethyl ketone and cellosolve acetate are, actually, the preferred sol¬ vents since they appropriately cαnbine good solvency pcwer and accept- able rates of evaporation frαn the freshly coated sheath.

In the freshly prepared diisoeyanate capped prepolymer solution, the concentration of free isoeyanate groups can be analyzed by a me¬ thod derived from the method described above for determining the free OH of the polyester diol. For this analysis, the isoeyanate capped material is weighed exactly (m gram) and a known excess of dibutyl¬ amine solution (in toluene) is added. A control blank with solvent only is made similarly, after which both samples are heated for some time and, after eooling, back titration with normal HC1 is carried out (indicator: bromocresol green) . The percent NGO is then obtain- ed as follσws (x πmole of HC1 N used) :

% NGO = «2L (MW _NC0 - 42)

In the present invention, with polymer to solvent weight ratios cαnprised between 0.25 and 1, the percent NCO varied fron about 0.35 to 5.2% which means that, in principle, the % free NCO of the undis- solved capped prepolymer was about 1.5% to 5.2% by weight depending on the case. Once the isoeyanate capped prepolymer is ready, it can be used for coating catgut filaments by using the coating methods known, in general, by people skilled in the art. Such means include immersion coating, spraying or die coating. A preferred method will be descri¬ bed hereinafter in the Examples. The useful viscosities of the ooat- ing solution will be determined by a number of factors such as coat¬ ing rate, desired coating thickness, type of polymer free isoeyana¬ te concentration properties to be given to the coating, etc. In

general, proper viscosity values will be obtained by adjusting the polymer to solvent ratio according to the needs. Viscosity values at ro temperature of fro 50 cP to 200 cP are generally suitable for coating thicknesses ranging fron about 1.2 rnι to 40-50 j . 5 When the filament has been coated with the isoeyanate capped polymer, it is allowed to stand for some time to harden in air. During this period, moisture curing cecurs, as mentioned hereinabove, which imparts to the sheath its final surface and body properties: soft- ness, slickness, flexibility and modulus. This curing can be carried _Q out in ordinary atmosphere at room temperature or it can be acceler- ated in a moisture oven between, say 30 and 70°C. If desired, the coating can be repeated for increasing the sheath thickness or for masking the pinholes which may have oecured in the first coating. Probabilities that two pinholes superimpose are, indeed, negligible. 5 When the sheathed catgut is finally ready it is sterilized and packed, either dry or in alcohol containing wrappers for being used in surgery.

Thus, the present invention effectively provides a suture fila¬ ment with the following useful properties: 0 a) It comprises a sleeving that is biodegradable at a rate govern- ed by the coating conditions and the type of polymer used and, de¬ pending on the needs, fron about 10 hrs to 3-4 weeks. b) It has a h nogeneous pinhole free layer which protects the core from rapid enzymatic degradation. c) It has an appropriate sliek surface which enables quick and easy suture work and, si ultaneously, prevents knot slippage. d) It is simple and cheap to manufacture. e) The sleeve has tensile properties under stress c nparable with that of the underlying catgut and it does not break when the fila- ment is stretched. Indeed, the coating polymer is slightly more ex- tensible than the oore of collagen. f) It adheres well to the catgut. g) It minimizes tissue reactions after stitching as compared with the effect of normal catgut (chremium treated or not) .

Reduction to Practice (Industrial Applications)

The Examples that follow illustrate the invention in which refe- rence is made to the annexed drawing. 5 Fig. 1 of this drawing represents schematically a device for coating a catgut filament with the capped prepolymer of the inven¬ tion.

Fig. 2 is ^" a diagram showing the cαmpared resistance to icro- biological attack of three sheathed catgut filaments and one unsheath- X0 ed oontrol.

Example 1

There were ixed together under nitrogen 0.0315 mole of perfluoro-

15 adipie acid and 0.05 mole of 1,6-hexanediol. The mixture was kept 1.5 hrs at 110°C and 0.005g of dry AlCl- were added (esterification catalyst) and thoroughly mixed . after which reduced pressure (0.005 Torr) was applied for evaporating the water formed by σondensation. The mixture was finally heated 4 hrs at 200°C and 1 hr at 235°C after 0 which it was cooled under dry nitrogen and stored as like. The pro- duct was a viscous liquid. Five grams of this prepolymer diol were dissolved with 7.5 ml of THF and 0.5 g of tolylene diisoeyanate was added under stirring. After a few minutes rest at room temperature, ^' a polished and untanned catgut filament 0.3 mm thick was coated by 5 dipping into the viscous fluid follσwed by draining. The solvent was removed with an air jet at 60°C after which .the coated filament was allowed to eure by staying 24 hrs in air at 60°C. For measuring the sheath thickness, the same operations were performed but adding to the polymer solution a small amount of methylene blue. Thus, the coat- 0 ing thickness could thereafter be measured on a microtome cut sect- ion of the wire under the microscope by observing the width of the colored circling area. It was found that the coating was about 10 micron thick.

The non-colored coated catgut was tested for degradation on rats 5 as follows: Sheathed and unsheathed control filament lenghts (about 4 cm) were stitched under the skin of a series of experi ental rats (Sprague Dawley) . The controls were chrαnium treated. After test pe-

riods of respectively two, seven and fifteen days scme of the test and control rats were sacrificed and the wounds were examined histo- logically. It was found that the filaments protected according to the present Example stayed virtually unattacked after two and seven days and that tissue swelling and inflaming was negligible and mark- edly less than that observed around the chrαnated catgut filaments after two days. Also, after fifteen days, the sheath of the filaments coated according to the present Example were only partly attacked whereas the control filaments were very strongly degraded.

Example 2

Fifty g (0.373 mole) of 2,2-oxydiacetic acid and 44.5 g (0.391 mole) of diethyleηe glycol were mixed with 0.05 g of A1C1-. and the whole was heated according to the following program:

Time (hrs) Temperature (°C) Pressure (Torr)

1 110 760

0.5 135 760

1 135 20

1 135 0.05

4 200 0.05

1 235 0.05

The resulting polyester was allcwed to oool and was dissolved in dry Chloroform, then it was filtered on glass frit to remove the cata¬ lyst. The solution was evaporated under vaccum and yielded a color- less waxy material. The No _QH value was analyzed as described here- inbefore and found to be 4,650 in rather close agreement with theory. The polyester diol was diluted with methyl ethyl ketone (MEK) or cello¬ solve acetate to ake a 52.5% by weight solution. This was stored äs the stock solution.

Aliquots of 3 g (0.0035 mole) of the above polyester-glycol solu- tions were further diluted with 3 g of MEK (or cellosolve acetate) to produce Solutions at 26.3% by weight and to each of the Solutions were added a quantity of TDI calculated for having the following mole

ratios diisocyanate/diol : a) 1.1; b) 1.3; c) 1.5. To the above Sol¬ utions were further added 0.25% by weight of polymer solids of DABCO catalyst. The obtained Solutions of NCO capped polyester glycol were then used for coating catgut filaments using the device schematical- ly pictured on Fig. 1. This device comprises a cup 1 of stainless steel or any other inert material in the bottom of which a tiny hole 2 has been drilled. The cup is filled with the polymer solution 3 and the diameter of the hole is adapted for enabling a catgut fila¬ ment 4 to freely pass therethrough but small enough to substantial- ly prevent the polymer solution to drift around the filament through the hole. The catgut 4 is attached to a pulling wire 5 which passes over pulley 6 and can be wound on a roll 7 by means of drive not re- presented here (motor or crank) . For setting the device into Opera¬ tion, the catgut filament is threaded frαn the bottom through the hole, crimped around the wire 5 and the whole is drawn at constant speed (1 αn/sec in the present Example) until the whole length of the filament has been coated by passing through the cup. The Over¬ all dimensions of the device are adjusted for having the coated fi¬ lament to just overhang frcm roller 6 without touching it not to dis- turb the freshly applied coating. Then, the filament is unerimped and allowed to stand in moist air for hardening. In the present Example this period was about 48 hrs. After this period, the sheath was considered to be fully cured and the surface was s ooth and glossy. Hσwever, tests made αi films cast on glass plates with the same polymer showed the absenee of free NCO groups by I.R. speetro- etry after already 12 hrs at 40°C p^ at 2240cm ^"1 ) . It is interest- ing to note that the coated catgut filaments thus prepared were sub¬ stantially transparent whereas non σoated catgut is cpaque. This feat- ure which probably results fron a proper matching of the refraction indexes of both materials is σommercially attractive.

Regarding nσw the differences obtained with coatings of types a) , b) and c) above, it was found that polymerization was faster with the coatings having the highest NCO/CH ratio; not much difference was found in the final properties, however, the c) sample being sαne- what more rigid than the other samples but with not much significan- ce. All samples had very good handling properties for stitching, ha¬ ving no knot slippage and well accαnodating catgut swelling when in

O.V.

-fy_. WiP

contact with aqueous fluids. No break of the sheath was experienced during manipulations. The thickness of the coatings, as measured as described above, was in the ränge of 2-5 m.

5 Example 3

A series of 6 polyester-diols (respectively, A to F) were pre- pared fron diethylene glycol and oxalic acid for samples A to C and ^" 2,2-oxydiacetic acid for samples D to F but changing the mole ratio 0 diacid to diol in Order to obtain different molecular weights for the polyesters. The reaction conditions were that described in Exam¬ ple 2 and the following Table I provides data on the various samples including the values for n (see the introduction) frαn which the COOH/OH ratios used were calculated and the experimental n calcul- _5 ated, as already described, fron the molecular weight determined ex- perimentally.

TABLE I

0 Sample diacid n (theory) NO MW n (frαn No)

- OH OH

A oxalic 1.2 390 290 1.04 B 4 146 765 3.9 C 5 135 830 4.3 5 D oxydiacetic 18 26.4 4250 16.7

E 14 34.3 3260 12.7 F 3 173 650 2.3

0 The polyester diols A to F were all diluted to 26.3% with MEK (or isobutyl ethyl ketone) and an amount of TDI was added to have an iso¬ eyanate prepolymer solution with an NGO/OH ratio of 1.5 together with 0.25% of DABCO. Then the Solutions were used to coat catgut filaments of diameter 0.5 mm with the device described in Example 2 at a speed 5 of 1 cm sec. After curing the coated filaments were tested for their handling properties by pulling, stitching and knotting. It was found that samples A, B, C and F were all right while samples D and E

sleeves had too much elasticity for proper handling this being due to using the higher molecular weight polyesterdiols. Thus, other samples were prepared similar to D and E but replacing 0.05 equiva- lent of the diethylene glycol by 0.05 equivalent of trimethylolpro- pane. When capping such modified polyester-glycols with TDI as des¬ cribed above and coating catgut filaments therewith, sleevings were obtained with much reduced elasticity due to the introduction of cross- -linking.

Example 4

An isoeyanate capped polymer solution was prepared correspond¬ ing exactly to sample F of the previous Example and, after the add- ition of 0.5% of bromocresol blue, it was used to ooat catgut fila- ment (1.4 m lengths) of various grades (No 1-0, 2-0 and 3-0 corres¬ ponding to thicknesses of 0.40, 0.35 and 0.30 mm, respectively) . The coating speed was 1 cm/sec. In seme cases the filaments were coat¬ ed twice (2 passes) after an intermediate curing interval of 30 min. During the interval the isoeyanate Solutions were kept away frem moist- ure to avoid premature polymerization.

After final curing the coating thicknesses were measured both by weighing (the bare catgut had been previously dried over P- _j O,- to constant weight) and under the microscope as described hereinbefore. The results are shown below:

Filament No Nb. of passes Coating thickness (um)

1-0 1 2

1-0 2 6

22--00 11 1.5

3-0 1 1.2

3-0 2 4.5

These data shew that the second ooating Operation more than doubles the sheath coating thickness. The advantage of a double coating is to efficiently mask s ne pinholes which might have possibly formed in the first coating and which would, otherwise, constitu attack-

ing sites for the catgut core.

Example 5

5 Three samples of sheathed catgut C, F and G were prepared by the technique described in Example 2. Samples C and F were identi- cal with the corresponding samples C and F of said Example and sam¬ ple G was prepared frαn a si ilar polymer but using glutaric acid as the diacid, the n value for the intermediate polyester glycol being O about 4. The sheath thickness was about 6-8 μa. and was ccmposed of a double layer ooating (2 passes) .

The above three samples were subjected to enzyme degradation testing, together with an uneoated control, as follows:

Prior to coating, a 60 cm long segment of catgut filament (No.5, 5 0.5 mm) was eoiled in a Petri box and wetted with 3 ml of an alco- hol-water buffer at pi 7.5 (0.05 M phosphate + 0.9% NaCl) . After 20 min. , the buffer was discarded and there was added into the box a mixture of 1 ml of the buffer and 10 ml of aqueous Na ¹²⁵ i (activity 1 mCurie) . After agitating for 20 min, there were added 20 jl of chlor- 0 a ine T solution (lOQug) , this solution being of 50 mg/1 of chlor- amine T in the above NaCl/phosphate buffer. After further 20 min of agitation, there were successively added 0.1 ml of a K^S^O _c solution

(at 1.334 g 1 in the NaCl/phosphate buffer), then 0.5 ml of a 0.4

M KI solution in the same buffer and, finally, 0.5 ml of tert.BuOH. After half an hour, the liquid was removed and the filament was tho- roughly rinsed with several portions of 5-6 ml each of the above al- cohol-water buffer. Then, the catgut was also rinsed with the norm¬ al sterilizing solution for preserving catgut sutures. The length of filament was dried and cut into 4 parts, three of them being coat- ed, including the ends, for making samples C, F and G as said abo¬ ve and the fourth remaining uneoated (control) . The samples were then allcwed to stand in pH 7.4 buffer for 48 hrs until no further radio- active iodine was liberated. Measurements of radioactivity were per- formed using a Gamma Radio Spectometer (Packard odel 300) . In such spectr neter the counting of the total sample (irrespective of its size) is easured. Degradation tests on the samples were performed as follows:

Each filament was placed in a test tube containing 5 ml of a sol¬ ution at pH 7.4 (0.07 M phosphate buffer) containing NaCl (0.08 M) , CaCl ₂ (0.01 M) , NaN ₃ (0.2 mg/ml), streptαnycine Sulfate (0.05 mg/ml) and collagenase (0.002 mg/ml). One hundred ιL aliquots were remov- ed at time intervals for radio-activity counting of the dissolved - ^Lji " ^J I thus liberated. The measurements were carried out first after hourly periods, thereafter fron dayly periods. Each day, the enzyme Solutions were replaced by identical fresh Solutions. The results are shown on the graph of Fig. 2. This graph plots, against time, the iodine liberated as a percent of the total iodine initially pre¬ sent. The obtained curves are labelled as the sample they illustrate. They shcw that the oxalic based polymer provided about 10-12 hrs of effective proteetion after which the core became subjected to enzy- matic attack (curve C is about parallel to the control curve) . Attack on sample F was much slower, thus providing adequate proteetion for about two weeks. Sample G (glutaric acid) was fairly inert during the corresponding period.

Example 6

Example 2 was repeated but using ^" 75.66 g (0.380 mole) of tetra- ethylene glycol instead of the diethylene glycol and 0.015 mole of glycerol. The polyester thus obtained was a waxy solid which was treat- ed with 1.5 equivalent of tolylene diisoeyanate in DMF to obtain a 18% solids solution of isoeyanate prepolymer. This solution was used to coat catgut lengths as described in Example 2. The coated fila¬ ments were allcwed to dry in air at room temperature for 48 hrs after which they were dipped into water and allαwed to stay for 12 hrs for eliminating all traces of EMF. After drying, the filaments had a smooth 10-12^ coating and had good handling properties.

7 VΓΪPO