Morphine Promotes Colonization of Anastomotic Tissues with Collagenase - Producing Enterococcus faecalis and Causes Leak

Abstract

Background: Despite ever more powerful antibiotics, newer surgical techniques, and enhanced recovery programs, anastomotic leaks remain a clear and present danger to patients. Previous work from our laboratory suggests that anastomotic leakage may be caused by Enterococcus faecalis strains that express a high collagenase phenotype (i.e., collagenolytic). Yet the mechanisms by which the practice of surgery shifts or selects for collagenolytic phenotypes to colonize anastomotic tissues remain unknown.

Methods: Here, we hypothesized that morphine, an analgesic agent universally used in gastrointestinal surgery, promotes tissue colonization with collagenolytic E. faecalis and causes anastomotic leak. To test this, rats were administered morphine in a chronic release form as would occur during routine surgery or vehicle. Rats were observed for 6 days and then underwent exploratory laparotomy for anastomotic inspection and tissue harvest for microbial analysis. These results provide further rationale to enhanced recovery after surgery (i.e., ERAS) programs that suggest limiting or avoiding the use of opioids in gastrointestinal surgery.

Results: Results demonstrated that compared to placebo-treated rats, morphine-treated rats demonstrated markedly impaired anastomotic healing and gross leaks that correlated with the presence of high collagenase-producing E. faecalis adherent to anastomotic tissues. To determine the direct role of morphine on this response, various isolates of E. faecalis from the rats were exposed to morphine and their collagenase activity and adherence capacity determined in vitro. Morphine increased both the adhesiveness and collagenase production of four strains of E. faecalis harvested from anastomotic tissues, two that were low collagenase producers at baseline, and two that were high collagenase producers at baseline.

Conclusion: These results provide further rationale to enhanced recovery after surgery (i.e., ERAS) programs that suggest limiting or avoiding the use of opioids in gastrointestinal surgery.

Keywords: Anastomotic leak; Enterococcus faecalis; Morphine.

Fig. 1

Morphine exposure worsens anastomotic healing in rats. (a–f), Images defining anastomotic healing score (AHS). a AHS0, an intact anastomosis (shown by black arrow) with no adhesions adherent or adjacent to the anastomosis. b AHS1, characterized by significant adhesions of the surrounding omentum shown by purple arrow. c AHS2, characterized by dense and highly thickened omental adhesions. Black arrow shows the site of anastomosis that is underneath the omental adhesion shown by purple arrow. d AHS3 displaying a wide and hard abdominal phlegmon around the anastomosis. Black arrow shows the site of anastomosis that is within an abdominal phlegmon formed by omental and small bowel adhesions shown by purple arrow. e AHS4 demonstrates a more severe abdominal phlegmon compared to AHS3 with gross abdominal contamination, pus, and a clear area of anastomotic dehiscence. f Internal (mucosal) view of a typical AHS3 or AHS4 level anastomotic leak via × 63 light microscopy. Left arrow depicts a healed anastomotic tissue line. Right arrow shows separation and full thickness disruption (leak) of anastomotic tissue line. g The distribution of rats treated with placebo (n = 30) versus morphine (n = 31) by AHS. h Cumulative Anastomotic Healing Score calculated as the weighted arithmetic mean AHS of all the rats within each group $\left({\sum }_{i=0}^{i=4}\frac{n_{\text{AHS}}}{n_{\text{Total}}}\times {\text{AHS}}_i\right)$. *p < 0.001 (Student’s t test)

Fig. 2

Colon surgery shifts the tissue-associated microbiota at anastomotic sites. a Principal coordinate analysis of preoperative (POD0) and postoperative (POD6) anastomotic tissue-associated microbiomes ordained by unweighted UniFrac distances. b Comparative analysis of bacterial relative abundance (genus level) in anastomotic tissues swabs and homogenized tissue at POD0 and POD6. c Shannon index alpha-diversity demonstrating colon surgery-induced decrease of microbiome diversity at POD6. Error bars = ±2 SE

Fig. 3

Relative abundance of adherent Enterococcus and frequency of culturing of E2 phenotype E. faecalis correlate to the increase in the anastomotic healing score. a 16S rRNA analysis of relative abundance of Enterococcus in adherent bacteria (anastomotic tissue swabs) in the correlation to the anastomotic healing score in the morphine-treated group. b Simple regression analysis of mean values of relative abundance of Enterococcus in groups of rats with different AHSs to anastomotic healing scores. c Frequency of anastomotic samples cultured positive for E. faecalis E2 phenotype in the group of 18 morphine-treated rats in the correlation to anastomotic healing score (*p < 0.005)

Fig. 4

Morphine induces collagenase production and adherence of E. faecalis. a–c The effect of morphine, 40 μg/ml, on the collagenase activity in E. faecalis isolates showing collagenase activity normalized to cell density (a), total collagenase (b), and cell density (c). d–f Effect of morphine on adhesiveness of E. faecalis showing adherence normalized to cell density (d) growth, total adherence (e), and cell density (f). n = 3, p < 0.01, results reproduced in three independent experiments