Original Articles Perioperative Matrix Metalloproteinase Inhibition Therapy Does Not Impair Wound or Anastomotic Healing James H. Balcom IV, M.D., Tobias Keck, M.D., Andrew L. Warshaw, M.D., Bozena Antoniu, M.S., Gregory Y. Lauwers, M.D., Carlos Fernández-del Castillo, M.D.

Matrix metalloproteinases (MMPs) catalyze the degradation of collagen and extracellular matrix. They play a role in pathologic states including malignancy, in which they facilitate invasion and metastasis. MMP inhibition has been shown to block neoplastic invasion and improve survival in animal models of malignancy. Concern about the effects of MMP inhibitors on wound and anastomotic healing may limit their potential use in the perioperative period to prevent local and systemic showering of cancer cells from surgical manipulation. We sought to assess the safety of perioperative administration of an MMP inhibitor (BB-94) with respect to skin and bowel healing in a rat model. Absorption of BB-94 was confirmed through high-pressure liquid chromatography and mass spectroscopy of sera from treated animals. Bowel bursting pressure in all animals increased almost 10-fold between 4 and 14 days. Two-way analysis of variance showed no significant difference in bowel bursting pressure between control and treatment animals over time. There was a significant increase in the collagen content of skin specimens of all animals combined between 4 and 28 days. Similarly, all animals showed an increase in bowel collagen between 4 and 28 days. There was no significant difference in skin or bowel collagen concentrations between control and treatment animals over time. Perioperative treatment with MMP inhibition does not impair wound or enteric healing in a rat model of laparotomy and small bowel resection. MMP inhibitors are safe for use as adjuvant therapy after resection for cancer. ( J GASTROINTEST SURG 2002;6:488–495.) © 2002 The Society for Surgery of the Alimentary Tract, Inc. KEY WORDS: Matrix metalloproteinase, wound healing, anastomosis, collagen, surgical oncology

Adhesion, migration, and subsequent function of cellular elements in mammalian tissues are intimately dependent on the content and function of the extracellular matrix in which they are found. Physiologic and pathologic processes, such as wound healing and neoplastic progression, influence and are affected by changes in the extracellular matrix.1,2 The composition of the extracellular matrix is in constant flux as collagen and other structural proteins are synthesized, deposited, and degraded.1 Matrix metalloproteinases (MMPs) are a group of proteolytic enzymes that function to degrade extracellular matrix. The MMP family of enzymes includes, among others, collagenases (represented by MMP-1),

gelatinases (represented by MMP-2 and MMP-9), and stromelysins (represented by MMP-3).3 MMP activity has been described in a variety of processes including inflammation, angiogenesis, wound healing, and cancer.2,4–6 There is now considerable evidence that MMPs may have a vital role in the malignant invasion of normal tissues by local extension and metastasis.2,7,8 Using a nude mouse model of metastatic pancreatic ductal adenocarcinoma, we have shown that MMP inhibition with the drug BB-94 (batimastat) decreases tumor load and improves survival. Our studies also suggest that this strategy may be particularly useful in the perioperative setting of tumor extirpation, where MMP inhibition may pre-

Presented at the Sixteenth Annual SSAT/Ross Residents and Fellows Research Conference, May 19, 2001; and at the Forty-Second Annual Meeting of The Society for Surgery of the Alimentary Tract, Atlanta, Georgia, May 20–23, 2001 (poster presentation). From the Departments of Surgery (J.H.B., T.K., A.L.W., B.A., C.F.C.) and Pathology (G.Y.L.), Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts. Supported by the Edward D. Churchill, M.D. Resident Research Fellowship, Harvard Medical School, Boston, Massachusetts. Reprint requests: Carlos Fernández-del Castillo, M.D., Massachusetts General Hospital—WACC 336, 55 Fruit St., Boston, MA 02114. e-mail: [email protected]

488

© 2002 The Society for Surgery of the Alimentary Tract, Inc. Published by Elsevier Science Inc.

1091-255X/02/$—see front matter PII: S 1 0 9 1 - 2 5 5 X ( 0 2 ) 0 0015- X

Vol. 6, No. 3 2002

vent showering of malignant cells into the local and systemic circulation and may prevent implantation of those same cells.9 The role and influence of MMPs in wound healing has been the subject of intense investigation for several years. There is evidence from numerous studies that MMPs are crucial in the coordination of wound healing and in the long-term extracellular matrix remodeling that occurs in its later stages.1,3,4,6,10–15 Interestingly, there is evidence that MMPs and their endogenous inhibitors, tissue inhibitors of metalloproteinases, may have complex roles in wound healing, which amounts to a net increase in collagen and other extracellular matrix in some circumstances and a net decrease in matrix components in others.1,8,12,14 Few experiments have included relevant measurements of mechanical wound strength in their assessment of the effect of MMPs on healing, particularly enteric healing. Without clear experimental evidence of an unequivocal positive or negative effect of MMPs on wound healing, the use of MMP inhibitors as perioperative adjuvants in surgical oncology has been understandably limited. For example, the ongoing clinical trial with the oral MMP inhibitor marimastat in pancreatic cancer requires a 6-week hiatus between resection and the initiation of drug treatment.7 The aim of this study was to determine the effect of BB-94 (batimastat), which is a broad-spectrum metalloproteinase inhibitor with proved in vitro and in vivo efficacy,9 on wound healing in a rat model of laparotomy and small bowel resection. MATERIAL AND METHODS Animals, Operative Technique, and Measurement of Bowel Bursting Pressure All animals used in these experiments were treated according to a protocol approved by the Subcommittee on Research Animal Care of the Massachusetts General Hospital in accordance with guidelines set forth in the “Guide for the Care and Use of Laboratory Animals” (NIH publication 86-23, 1986). Male Sprague-Dawley rats (n  75), weighing 150 to 250 g, were purchased from Charles River Laboratories (Wilmington, Massachusetts) and were fed standard dry chow and water ad libitum for 3 days. The animals were then fasted overnight and given only a solution of 0.225% normal saline with 5% sucrose by mouth. On the morning of operation, the animals were placed under general anesthesia with intramuscular ketamine, 44 mg/kg, and intraperitoneal pentobarbital, 20 mg/kg. Animals were then randomly assigned to one of three groups: (1) intraperitoneal BB-94, 40 mg/kg, in 20% dimethyl sulfoxide (DMSO;

Inhibition Does Not Impair Healing

489

Sigma, St. Louis, Missouri) for 3 days beginning on the day of operation; (2) intraperitoneal BB-94, 40 mg/kg, in 20% DMSO for 14 days beginning on the day of operation; or (3) intraperitoneal solution of 20% DMSO in saline for 3 or 14 days beginning on the day of operation. BB-94 was generously provided by British Biotech (Oxford, United Kingdom). Because of its limited water solubility, BB-94 was prepared for injection in a sterile solution of 20% DMSO in normal saline. The resulting suspension was homogeneous, and we were able to use it for injection without difficulty. Approximately 30 minutes after the drug or carrier was injected, a midline laparotomy was performed and a loop of distal ileum elevated into the wound. A 3 cm segment of ileum was resected, and the bowel was reanastomosed in end-to-end fashion with five interrupted, inverting 5-0 polyglycolic acid sutures. The abdomen was closed in two layers with running 5-0 polyglycolic acid sutures, and the animals were allowed to recover. The animals were given the saline and sucrose solution, which was described previously, on postoperative day 1, and thereafter they were given standard dry chow and water ad libitum. Intraperitoneal injections of drug or carrier were administered according to the schedule outlined previously, and the animals were killed by carbon dioxide inhalation on postoperative day 4, 10, 14, 28, or 49 (n  5 animals/ group). Bowel bursting pressure (BBP) was determined at that point by insufflating the anastomosed segment of bowel with air in parallel connection to a mercury sphygmomanometer column. The BBP was recorded as the pressure required to cause bowel disruption at the anastomosis. For hydroxyproline and collagen determinations, a 6 mm punch biopsy was used to obtain samples of skin from the area of the healed skin incision and intestine from the area of the bowel anastomosis. Separate specimens from both locations were (1) snap-frozen in liquid N2 and stored at 80 C and (2) fixed in formalin for preparation of paraffin-embedded tissue blocks. Reverse-Phase High-Pressure Liquid Chromatography and Mass Spectroscopy Although a previous study has shown that BB-94 can be absorbed based on evidence from murine intraperitoneal injections,16 we sought to confirm this. In four rats, 8 ml of blood was collected by cardiac puncture on day 14 of BB-94 treatment, just before the animals were killed. The separated plasma was combined into a single aliquot, and the plasma proteins were removed by methanol precipitation according to the method described by Wang, et al.16

490

Journal of Gastrointestinal Surgery

Balcom et al.

Briefly, the methanol-precipitated samples were lyophilized and reconstituted in 60% acetonitrile and submitted for reverse-phase high-pressure liquid chromatography (RP-HPLC). All RP-HPLC and mass spectroscopy analyses were performed at Analytical Technology Services, Watertown, Massachusetts. The methanol-precipitated plasma was then analyzed by means of RP-HPLC using an RP–C-8 analytical column on a Varian reverse-phase high-pressure liquid chromatograph (Varian, Inc., Palo Alto, California). The retention time of BB-94 was determined by injection of 25 g of pure BB-94 and BB-94 “spiked” blood at different concentrations. An optimal elution gradient of 0.1% trifluoroacetic acid (TFA) and 99% acetonitrile at 1 ml/min for 20 minutes revealed desorbtion of BB-94 at 13.74 minutes. The methanol-precipitated plasma was collected in fractions every 30 seconds from 13 to 15 minutes and submitted for mass spectroscopy. Mass spectroscopy was performed using a Voyager Elite mass spectrometer (PerSeptive Biosystems, Inc., Framingham, Massachusetts) and workstation. The 13.5-minute fraction from the methanol-precipitated plasma demonstrated a single, large peak at 478 atomic mass units (amu), which corresponds to the peak seen on mass spectroscopy of BB-94 “spiked” blood. Hydroxyproline and Collagen Analysis All procedures used in the determination of hydroxyproline and collagen content were based on the method described by Reddy and Enwemeka.17 Briefly, the method involves the following three steps: (1) hydrolysis of the skin and bowel specimens; (2) oxidation of the liberated hydroxyproline; and (3) development of a chromophore with the oxidized hydroxyproline. The samples were air dried, weighed, and autoclaved at 120 C for 20 minutes in a 2 mol/L NaOH solution to liberate hydroxyproline. The samples were then mixed with 0.056 mol/L chloramine T (Sigma), which results in hydroxyproline oxidation and production of a pyrrole. Addition of 1 mol/L Ehrlich’s reagent (Sigma) to the oxidized hydroxyproline results in the generation of a chromophore. The absorbance of the chromophore can be measured at 550 nm. Known concentrations of hydroxyproline stock solutions were subjected to the process described above, and the resulting absorbance measurements were used to create a linear standard curve. Absorbance measurements from each sample, prepared as described above, were then used to obtain hydroxyproline concentrations from the standard curve. Since hydroxyproline expression is essentially limited to col-

lagen, we assumed that collagen consisted of 12.5% hydroxyproline to calculate the collagen content,17,18 which was normalized for tissue weight in each sample. Trichrome Staining of Paraffin-Embedded Tissue Slides were prepared from the previously collected paraffin-embedded skin and bowel specimens and stained with a standard Masson’s trichrome staining technique. The slides were then reviewed in blinded fashion by a pathologist (G.Y.L.). A semiquantitative scoring system was used to assess the amount of collagen present in skin or bowel wounds: 0  no organized collagen deposition in the dermis or submucosa; 1  minimal collagen deposition; 2  moderate collagen deposition; 3  marked collagen deposition. Statistical Analysis To formulate a comprehensive interpretation of potential differences among the three experimental groups over the entire duration of the study, comparisons of the BBP values, skin collagen concentrations, bowel collagen concentrations, and histologic collagen score were conducted using a two-way analysis of variance (ANOVA) technique. Comparisons of BBP values, skin collagen concentrations, and bowel collagen concentrations at specific time points in the study were performed using Student’s t test. Continuous data are presented as the mean  SEM. Statistical significance was defined as P  0.05. All statistical analyses were performed with the assistance of GraphPad Prism version 3.00 for Windows, GraphPad Software, San Diego, California.

RESULTS BB-94 treatment was well tolerated. The animals treated with BB-94 did not experience weight loss or increased mortality (data not shown). RP-HPLC of pure BB-94 in acetonitrile and the methanol-precipitated plasma was performed as described previously. Because of the broad nature of the peak at 220 nm seen at 13.74 minutes of extraction time, multiple fractions from 13 to 15 minutes were collected during the separation (data not shown). Mass spectroscopy of BB-94 “spiked” blood and the 13.5-minute HPLC-separated fraction yielded dominant peaks clearly identified at 478 amu (data not shown); these peaks corresponded to the published molecular weight of 478 amu for BB-94.19

Vol. 6, No. 3 2002

Inhibition Does Not Impair Healing

491

Fig. 1. A, Significant increase in BBP between 4 and 14 days for all animals killed at these time points (increase from 28  2 mm Hg at 4 days to 249  16 mm Hg at 14 days, P  0.001; *P  0.05). B, Changes in BBP for animals in the three experimental groups over time (P  0.53).

BBP values are shown in Fig. 1. Fig. 1, A demonstrates the near 10-fold increase in BBP between days 4 and 14 (increase from 28  2 mm Hg at 4 days to 249  and 16 mm Hg at 14 days, P  0.001). Fig. 1, B shows the changes in BBP for animals in the three experimental groups over time. Two-way ANOVA showed no significant differences among control animals, animals treated with BB-94 for 3 days, and animals treated with BB-94 for 14 days (P  0.53). Skin wound collagen concentrations are shown in Fig. 2. There was a significant increase in skin collagen content between 4 and 28 days for all animals combined (9.0  0.8 mg/g tissue at 4 days vs. 11.7  0.8 mg/g tissue at 28 days, P  0.05) (Fig. 2, A), but there was no significant difference in skin wound collagen concentrations among the three experimental groups over time (Fig. 2, B; P  0.94). Fig. 3 shows the results of collagen concentration determinations in the bowel anastomoses of the experimental animals. Bowel collagen concentrations increased significantly between 4 and 28 days for all animals killed at these time points (2.0  0.2 mg/g tissue at 4 days vs. 4.8  0.6 mg/g tissue at 28 days, P  0.001) (Fig. 3, A), but bowel collagen content did not change significantly over time among the three experimental groups by two-way ANOVA (Fig. 3, B; P  0.63).

Results of histologic collagen trichrome staining of rat skin and bowel anastomoses are illustrated in Fig. 4. There were no significant differences in the histologic collagen scores in rat skin or bowel among the three experimental groups over time (P  0.05 for both). Typical trichrome stain results for early and late bowel anastomoses in control and treatment animals are presented in Fig. 5. Note that the early and late scars show the presence of similar amounts of collagen in control and treatment animals.

DISCUSSION The effects of MMP inhibition on wound healing and extracellular matrix turnover are complex and are not fully understood.1,3,5,20 Different MMPs appear to be expressed at different time points in the healing process. For example, whereas MMP-1 and MMP-9 appear to be upregulated early in the inflammatory and epithelial migration/proliferation periods of healing, MMP-2 seems to increase and remain elevated 5 days after wounding.1,3,20 Thus MMPs seem to have roles in both the acute phase of healing and the later stages of collagen and matrix remodeling. In addition, MMPs have proven effects on inflammatory cell migration and infiltration of freshly

492

Balcom et al.

Journal of Gastrointestinal Surgery

Fig. 2. A, Increase in skin collagen content between 4 and 28 days for all animals combined (9.0  0.8 mg/g tissue at 4 days vs. 11.7  0.8 mg/g tissue at 28 days, P  0.05). B, Skin wound collagen concentrations in the three experimental groups over time (P  0.94).

injured tissue and on angiogenesis,5,21,22 both of which also have implications for healing. The variety of influences on different aspects of healing requires that any study evaluating potentially deleterious effects of metalloproteinase inhibition include functional tests (such as BBP) in addition to morphologic studies. Perioperative treatment with BB-94 did not have a significant effect on BBP of the small bowel anastomosis in our study. The control BBP measured in our studies compares favorably with the BBP measured in the small and large bowel of operated and nonoperated rats in other published experiments; the range of bursting pressures reported in these other studies was 178 to 244 mm Hg for nonoperated bowel and 180 to 233 mm Hg for anastomosed bowel on postoperative day 7.23–25 The studies of Syk et al.11 and Witte et al.12 examined the effects of MMP inhibition on wound breaking strength in the skin and colon, respectively. Similar to the findings in our experiments, these other studies indicated that MMP inhibition did not impair wound breaking strength. In fact, the studies of Syk et al.11and Witte et al.12 found that the strength of the wound may be increased by MMP inhibition. This slight disparity may be explained by differences in the types of tissues analyzed, measurement techniques, and time

points of measurement. We chose to measure the anastomotic strength in the small intestine, whereas Syk et al.11 examined colonic wounds, and there is compelling evidence that the MMP profile in the small and large bowel may be quite different.3 Whereas Syk et al.11and Witte et al.12 used tensiometry to measure wound breaking strength, we used increasing intraluminal pressure to interrogate the repair, and this technique may better represent the physiologic demands of a bowel anastomosis. The other studies measured wound breaking strength at single time points in the relatively early stages of wound healing (day 10 in the skin study and days 1, 3, and 7 in the colon study).11,12 We measured and then analyzed BBP over a range of time periods, which concluded at 7 weeks, the time point classically associated with a maximal gain in wound tensile strength.26 The hydroxyproline and trichrome data from our studies provided us with information about the total collagen content of the skin and bowel anastomoses in the experimental animals. The collagen content of the small intestinal anastomoses in these experiments is similar to the collagen content determined in the small and large bowel of operated animals in other studies27–30; the range reported for operated bowel at the site of intestinal anastomosis, as determined by similar methods, is 2.01 to 2.36 mg collagen/g tis-

Vol. 6, No. 3 2002

Inhibition Does Not Impair Healing

493

Fig. 3. A, Bowel collagen concentration increase between 4 and 28 days for all animals killed at these time points (2.0  0.2 mg/g tissue at 4 days vs. 4.8  0.6 mg/g tissue at 28 days, P  0.001; *P  0.05). B, Bowel collagen content over time in the three experimental groups (P  0.63).

sue.30 As expected, the skin and bowel collagen concentrations increased from day 4 to day 28 when all animals were considered collectively. This stage in healing represents a relative peak in collagen deposition with a concomitant gain in wound tensile strength.26 Our study has shown that there was no significant change in the total collagen content of the

skin or bowel in rats treated with BB-94, 40 mg/kg, for 3 or 14 days over a 7-week experimental period. Other studies have produced similar results. Witte et al.12 showed that the collagen content in dorsal skin wounds in rats did not change significantly during treatment with a different MMP inhibitor, GM 6001, and Syk et al.11 showed that there was no dif-

Fig. 4. A, Histologic skin collagen scores in the three experimental groups over time (P  0.48). B, Histologic bowel collagen scores in the three experimental groups over time (P  0.39).

494

Journal of Gastrointestinal Surgery

Balcom et al.

Fig. 5. Collagen trichrome staining at the bowel anastomosis. A, Control rat at 4 days. B, BB-94–treated rat at 4 days. C, Control rat at 49 days. D, BB-94–treated rat at 49 days. Note the marked collagen deposition present in the scar at the bowel anastomosis of the control and BB-94–treated rats at 49 days (arrows).

ference in the hydroxyproline content of sponges placed in the wounds of rats on postoperative days 3 and 7 compared to control animals. In terms of a mechanistic explanation for these observations, the authors postulate that MMP inhibition leads to an enhancement of collagen maturity and cross-linking and/or some other modulation of the extracellular matrix that strengthens the wound; and Syk et al.11 pointed out that other studies have shown a beneficial effect for proteinase inhibition without a concomitant increase in collagen content.21,31 Our data would seem to corroborate these suggestions.

time points of measurement, this study along with other existing data suggests that the perioperative administration of therapeutic doses of MMP inhibitors is safe with regard to the maintenance of wound strength in oncologic resections. We thank British Biotech for providing the BB-94. We also thank the Boston Biomedical Research Institute and the Analytical Technology Services, Watertown, Massachusetts, for their assistance with the RP-HPLC and mass spectroscopy.

REFERENCES

CONCLUSION We have shown that perioperative MMP inhibition with BB-94 does not impair skin wound or enteric healing in a rat model of laparotomy and small bowel resection. We specifically chose the 40 mg/kg dosage and the 3- and 14-day treatment lengths because they have proven effectiveness in animal models of cancer inhibition.9,16 Therefore, regardless of differences in specific tissue types and techniques or

1. Soo C, Shaw WW, Zhang X, et al. Differential expression of matrix metalloproteinases and their tissue-derived inhibitors in cutaneous wound repair. Plast Reconstr Surg 2000; 105:638–647. 2. Heath EI, Grochow LB. Clinical potential of matrix metalloprotease inhibitors in cancer therapy. Drugs 2000;59: 1043–1055. 3. Seifert WF, Wobbes T, Hendriks T. Divergent patterns of matrix metalloproteinase activity during wound healing in ileum and colon. Gut 1996;39:114–119. 4. von Lampe B, Barthel B, Coupland SE, et al. Differential expression of matrix metalloproteinases and their tissue in-

Vol. 6, No. 3 2002

5.

6. 7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

hibitors in colon mucosa of patients with inflammatory bowel disease. Gut 2000;47:63–73. Raza SL, Cornelius LA. Matrix metalloproteinases: Proand anti-angiogenic activities. J Investig Dermatol Symp Proc 2000;5:47–54. Agren MS. Gelatinase activity during wound healing. Br J Dermatol 1994;131:634–640. Jones L, Ghaneh P, Humphreys M, Neoptolemos JP. The matrix metalloproteinases and their inhibitors in the treatment of pancreatic cancer. Ann N Y Acad Sci 1999;880: 288–307. Deryugina EI, Bourdon MA, Reisfeld RA, Strongin A. Remodeling of collagen matrix by human tumor cells requires activation and cell surface association of matrix metalloproteinase-2. Cancer Res 1998;58:3743–3750. Jimenez RE, Hartwig W, Antoniu BA, et al. Effect of matrix metalloproteinase inhibition on pancreatic cancer invasion and metastasis: An additive strategy for cancer control. Ann Surg 2000;231:644–654. Vaalamo M, Karjalainen-Lindsberg M, Puolakkainen P, et al. Distinct expression profiles of stromelysin-2 (MMP-10), collagenase-3 (MMP-13), macrophage metalloelastase (MMP12), and tissue inhibitor of metalloproteinases-3 (TIMP-3) in intestinal ulcerations. Am J Pathol 1998;152:1005–1014. Syk I, Agren MS, Adawi D, Jeppsson B. Inhibition of matrix metalloproteinases enhances breaking strength of colonic anastomoses in an experimental model. Br J Surg 2001;88: 228–234. Witte MB, Thornton FJ, Kiyama T, et al. Metalloproteinase inhibitors and wound healing: A novel enhancer of wound strength. Surgery 1998;124:464–470. Rechardt O, Elomaa O, Vaalamo M, et al. Stromelysin-2 is upregulated during normal wound repair and is induced by cytokines. J Invest Dermatol 2000;115:778–787. Colandrea T, Wang L, Wille J, et al. Epidermal expression of collagenase delays wound-healing in transgenic mice. J Invest Dermatol 1998;111:1029–1033. Okada A, Tomasetto C, Lutz Y, et al. Expression of matrix metalloproteinases during rat skin wound healing: Evidence that membrane type-1 matrix metalloproteinase is a stromal activator of pro-gelatinase A. J Cell Biol 1997;137:67–77. Wang X, Fu X, Brown PD, Crimmin MJ, Hoffman RM. Matrix metalloproteinase inhibitor BB-94 (batimastat) inhibits human colon tumor growth and spread in a patientlike orthotopic model in nude mice. Cancer Res 1994;54: 4726–4728. Reddy GK, Enwemeka CS. A simplified method for the

Inhibition Does Not Impair Healing

18.

19. 20.

21. 22.

23.

24.

25.

26.

27. 28.

29.

30.

31.

495

analysis of hydroxyproline in biological tissues. Clin Biochem 1996;29:225–229. Edwards CA, O’Brien WD. Modified assay for determination of hydroxyproline in a tissue hydrolyzate. Clin Chim Acta 1980;104:161–167. Wojtowicz-Praga SM, Dickson RB, Hawkins MJ. Matrix metalloproteinase inhibitors. Invest New Drugs 1997;15:61–75. Pilcher BK, Wang M, Qin X, et al. Role of matrix metalloproteinases and their inhibition in cutaneous wound healing and allergic contact hypersensitivity. Ann N Y Acad Sci 1999;878:12–24. Jonsson T, Hogstrom H. Neutrophil-dependent decrease in early wound margin strength. Arch Surg 1991;126:1423–1426. Schwartz JD, Shamamian P, Monea S, et al. Activation of tumor cell matrix metalloproteinase-2 by neutrophil proteinases requires expression of membrane-type 1 matrix metalloproteinase. Surgery 1998;124:232–238. Furst MB, Stromberg BV, Blatchford GJ, Christensen MA, Thorson AG. Colonic anastomoses: Bursting strength after corticosteriod treatment. Dis Colon Rectum 1994;37:12–15. Del Rio JV, Beck DE, Opelka FG. Chronic perioperative steroids and colonic anastomotic healing in rats. J Surg Res 1996;66:138–142. Fukuchi SG, Seeburger JL, Parquet G, Rolandelli RH. Influence of 5-fluorouracil on colonic healing and expression of transforming growth factor-beta 1. J Surg Res 1999;84: 121–126. Lawrence WT. Wound healing biology and its application to wound management. In O’Leary JP, ed. The Physiologic Basis of Surgery. Philadelphia: Williams & Wilkins, 1996, pp 118–140. Jonsson K, Jiborn H, Zederfeldt B. Collagen metabolism in small intestinal anastomosis. Am J Surg 1987;154:288–291. Bowers RL, Eeckhout C, Weisbrodt NW. Actomyosin, collagen, and cell hypertrophy in intestinal muscle after jejunoileal bypass. Am J Physiol 1986;250:70–75. Uden P, Blomquist P, Jiborn H, Zederfeldt B. Impact of long-term relative bowel rest on conditions for colonic surgery. Am J Surg 1988;156:381–385. Kiyama T, Onda M, Tokunaga A, et al. Effect of early postoperative feeding on the healing of colonic anastomoses in the presence of intra-abdominal sepsis in rats. Dis Colon Rectum 2000;43(Suppl 10):54–58. Tani T, Tsutamoto Y, Eguchi Y, et al. Protease inhibitor reduces loss of tensile strength in rat anastomosis with peritonitis. J Surg Res 2000;88:135–141.