Effects of Perioperative Antiinflammatory and Immunomodulating Therapy on Surgical Wound Healing

Anthony J. Busti, PharmD; Justin S. Hooper, PharmD; Christopher J. Amaya, PharmD; Salahuddin Kazi, MBBS


Pharmacotherapy. 2005;25(11):1566-1591. 

In This Article

Drugs Used in Various Inflammatory Disease States and Their Effects on Wound Healing

Corticosteroids affect the inflammatory process by stabilizing lysosomes within neutrophils, inducing antiinflammatory proteins, and inhibiting cytokine release and chemotaxis.[121,122] By reducing neutrophil chemotaxis, the host may be more susceptible to bacterial infection.[123] Collagen production, angiogenesis, and reepithelialization in dermal wounds may also be impaired.[124] In addition, fibroblast dysfunction occurs with the use of corticosteroids, which decreases wound tensile strength, an important component of the remodeling phase.[125,126] The clinical consequences of these effects include dehiscence of incision site, wound infection, and delayed open-wound healing.[114]

Table 4 summarizes animal and human studies performed to evaluate the effects of corticosteroids on wound healing.[126,127,128,129,130,131,132,133] In the first study, the effect of moderate-dose long-term corticosteroid administration on tensile strength of healing wounds was examined in rats.[126] The investigators divided 37 rats into five subgroups according to wound age and cortisol administration. Groups 1 and 2 included rats with 10-day-old wounds with or without cortisol administration, whereas groups 3 and 4 consisted of rats with 20-day-old wounds with or without cortisol administration. The fifth group consisted of intact rats serving as controls. Rats assigned to the corticosteroid group received cortisol 5 mg/kg subcutaneously for 42 days before wounding and until sacrifice. Forty-millimeter incisions were made on the backs of the rats and were immediately sutured closed. After sacrifice, 30-mm specimens were taken from the rats at sites adjacent to the wounds and were used to determine collagen content. Strips of each tissue sample were placed in a materials testing machine that loaded each specimen to breakage for determination of tensile strength as well as stress and strain values. The investigators found that samples taken from cortisol-treated rats were significantly thinner than those taken from intact rats, but collagen content was not affected. At day 20, cortisol-treated rats had skin wounds that demonstrated decreased failure energy (49.1 B1 3.9 vs 61.1 B1 4.4 N/mg/mm, p value not given) compared with tissue taken from controls. The authors concluded that moderate doses of corticosteroids given to rats over a long period mildly decrease the ability of wounds to withstand shear force, leading to dehiscence.

Another study used the rabbit osteotomy model to investigate the effects of systemic administration of corticosteroids on fracture healing.[127] Eighteen rabbits received either sterile saline (placebo) or prednisone 0.15 mg/kg/day subcutaneously for 2 months before osteotomy followed by 6 weeks after the procedure. The forelimbs of the rabbits were examined weekly for fracture union by radiography. At the end of 6 weeks, the rabbits were sacrificed, and dual-energy x-ray absorptiometry and mechanical testing were performed on each specimen. Investigators found that bone union occurred in only 3 of 20 limbs of rabbits that received prednisone, whereas 13 of 16 limbs of rabbits in the control group had fracture union (p<0.001). Bone mineral density and content at the osteotomy site and in adjacent bone were statistically lower in prednisone-treated rabbits compared with controls. The study implies that long-term corticosteroid administration may impair bone healing after surgical repair.

A similar study was performed in 20 rabbits that were administered cortisone acetate 6.25 mg/kg/day intramuscularly or an equivalent dosage of methylprednisolone 1 mg/kg/day intramuscularly, medroxyprogesterone 35 mg/kg/day intramuscularly, a combination of methylprednisolone and medroxyprogesterone at the same dosages, or placebo.[128] The rabbits were wounded on day 0, and the progress of wound contracture and maturation as well as epithelialization was recorded at prespecified intervals thereafter. In both the cortisone- and methylprednisolone-treated rabbits, an increase in the latent period before wound contracture was observed. Wound epithelialization was significantly delayed in rabbits receiving cortisone acetate when compared with controls (25 vs 20.5 days, p value not provided). Wound maturation was also delayed in rabbits that received methylprednisolone. These findings suggest that corticosteroids inhibit wound healing and may do so by mechanisms other than inhibition of inflammation.

Another group administered prednisone 2 mg/kg intramuscularly or isotonic saline (control) to a population of 72 rabbits that underwent mechanical dilatation and damage to the thoracic aorta.[129] After 2 weeks of prednisone administration, the rabbits were sacrificed and selected at random for determination of in vitro synthesis of [14]C-hydroxyproline (a major component of collagen) in the aorta in addition to collagen and other protein synthesis. The investigators found that the biosynthesis of soluble collagen was suppressed in the rabbits that received prednisone (6.6 B1 1.5% vs 10.0 B1 0.5%, p<0.01). This finding was substantiated by a smaller amount of DNA and RNA in the aortas of prednisone-treated rabbits when compared with rabbits that received saline injection (24.7 B1 3.5 vs 41.2 B1 5.2 B5g total DNA, p<0.02). In addition, prednisone administration appeared to inhibit aortic intimal thickness (24.6 B1 2.6 vs 8.1 B1 1.2 B5m, p<0.002). This finding showed that prednisone may exert an inhibitory effect on vascular connective tissue repair.

Vitamin A has been shown to reverse the effects of corticosteroids on tissue repair. A study sought to determine if this effect coincides with the reestablishment of inflammation and fibroplasia.[130] Rats, not designated as controls, received subcutaneous implantation of polyvinyl sponges followed by a single subcutaneous injection of either prednisolone acetate 8 mg (16 rats), vitamin A palmitate 5000 IU intramuscularly on the day of surgery and on days 2 and 4 (8 rats), or a single injection of both on the day of surgery (8 rats). At 3 and 7 days after implantation, three rats from each treatment group were sacrificed, and the sponges were removed and underwent histologic grading. A similar procedure was used to determine collagen accumulation on the implants. At day 3, the rats that were administered a corticosteroid were noted to have a significant absence of fibroblasts in surgical implants. Sponge disc implants from this same group of rats also had a decreased number of red blood cells (0 vs 0.71, p<0.05), polymorphonucleated cells (1.12 vs 2.56, p<0.05), and collagen fibers (1.75 vs. 2.29, p<0.05) compared with controls. Thus, the authors showed that glucocorticoids have an inhibitory action on the migration of inflammatory cells and fibroblasts as well as deposition of collagen fibers in an animal model.

One of the earliest reports on the effect of corticosteroids on wound healing in humans was made in 1965.[131] This investigator retrospectively examined the charts of 38 surgical patients who took corticosteroids preoperatively. Arbitrarily, he divided patients into groups according to the length of time they had received corticosteroids. Patients taking corticosteroids for at least 1 week were considered to be receiving "long-term" therapy. Most of the patients underwent splenectomy. The investigator found that 29% of all patients who received preoperative corticosteroids had complications (complete or partial disruption or prolonged discharge) with wound healing. No statistical analysis was performed, however. The investigator stated that there was no difference in complication rates based on the length of preoperative corticosteroid administration, but he did not account for confounding factors. This study represents one of the first known studies showing a link between corticosteroid administration and impaired surgical wound healing.

In another retrospective analysis in humans, authors analyzed the postoperative complication rate in patients with rheumatoid arthritis who underwent orthopedic surgery.[132] Of 111 total subjects, 49 had used corticosteroids (prednisolone 2.5-15 mg/day or its equivalent) within the previous 2 years. The authors found no difference between the times to complete wound healing in patients with rheumatoid arthritis compared with controls (16.6 B1 7.5 vs 15.2 B1 7.9 days, p>0.10). Compared with patients who did not receive corticosteroids, no difference was noted in wound healing among patients receiving corticosteroids (rates not given, p>0.10). However, the authors found that patients who had taken corticosteroids for more than 3 years took longer to heal after surgery (20.3 B1 11 vs 15.2 B1 4.9 days, p<0.05). Although these findings appear to be of clinical importance, it is difficult to assess the nonsignificance of the primary end point (days to complete wound healing) since no power analysis was done a priori.

Finally, in a retrospective study, authors reviewed the rates of fusion in 26 patients with rheumatoid arthritis who had either compression arthrodesis or internal fixation of the ankle.[133] Each group of patients was similar with regard to age and sex, and all of the patients had disease involvement in other joints. The authors found that four of seven ankles that failed to fuse were associated with infection. These patients were taking dosages of prednisone ranging from 6.5-40 mg/day (mean 17 mg/day), whereas patients who experienced failed fusion without infection were only taking an average of 6 mg/day. Although the study is limited by its small population, it is possible that higher dosages of prednisone may be a risk factor for failed fusion.

There is an emerging role for the use of other immunosuppressant drugs in the treatment of inflammatory disease states such as inflammatory bowel disease and rheumatoid arthritis. Tacrolimus was found to be useful in the treatment of DMARD-resistant rheumatoid arthritis, and its congener, sirolimus, has also shown similar promise.[134,135,136] Cyclosporine may be used for patients with inflammatory bowel disease and refractory fistulas; it has also been studied for use in patients with rheumatoid arthritis who have failed initial therapy.[137,138] Mycophenolate mofetil also may be a treatment option for rheumatoid arthritis.[139] Although these agents are traditionally used as immuno-suppressants after organ transplantation, their scope appears to be evolving to include treatment of other conditions such as rheumatoid arthritis and inflammatory bowel disease. Regardless of the clinical setting, practitioners should be mindful of their effects on wound healing in the perioperative period.

One group conducted a retrospective analysis of 158 adult patients undergoing kidney transplantation and receiving immunosuppression with either mycophenolate mofetil or sirolimus.[140] They showed that sirolimus was an independent risk factor for postoperative surgical complications and superficial healing problems. These results were confirmed in a prospective trial that showed an increase in superficial wound infections in patients receiving a sirolimus-based immunosuppressive regimen.[141] Similar results were found in rats that underwent sigmoidostomy and received either mycophenolate mofetil or placebo.[142] A significant decrease in healing and mechanical stability was found in the group that received mycophenolate mofetil. These findings underscore the importance of further research to determine the optimal time to resume therapy in the postoperative period in order to optimize healing and prevent disease flare-ups.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: