Persistent elevation of plasma vascular endothelial growth factor levels during the first month after minimally invasive colorectal resection

BACKGROUND

Elevations of plasma vascular endothelial growth factor (VEGF) have been noted early after colorectal resection. The duration of this increase is unknown. Because VEGF is a potent promoter of angiogenesis, which is critical to tumor growth, a sustained increase in blood VEGF levels after surgery may stimulate the growth of residual metastases early after surgery. This preliminary study aimed to determine VEGF levels during the first month after colorectal resection.

METHODS

Patients from three prospective studies that had late postoperative blood samples available comprised the study population. Demographic, perioperative, pathologic, and complication data were collected. Plasma samples were obtained preoperatively for all patients: on postoperative day (POD) 1 for most patients and at varying time points thereafter during the first month after surgery and beyond. Levels of VEGF were determined via enzyme-linked immunoassay (ELISA) and compared using Wilcoxon's matched pairs test. Because the numbers of specimens beyond POD 5 were limited, samples from 7-day time blocks were bundled and averaged to permit statistical analysis.

RESULTS

A total of 49 patients with cancer and 30 patients with benign indications, all of whom underwent minimally invasive colorectal resection, were assessed separately. With regard to the patients with cancer, the median preoperative plasma value was 150 pg/ml, and the peak postoperative median value for the POD 14 to 20 time block was 611.1 pg/ml. Furthermore, compared with the preoperative results, significant VEGF elevations were noted on POD 3 as well as during week 2 (POD 7-13), week 3 (POD 14-20), and week 4 (POD 21-27) (p < 0.05 for each). With regard to the benign patients, the median preoperative VEGF level was 112 pg/ml, and the peak postoperative value, 286 pg/ml, was noted during postoperative week 2. Significant elevations were noted on POD 3, and for weeks 2 and 3 as well as for POD 28 and later. Between 63% and 89% of the patients at each time point beyond POD 5 had elevated VEGF levels.

CONCLUSION

This preliminary study demonstrates that after minimally invasive colorectal resection for cancer, median VEGF levels are significantly elevated on POD 3 and remain increased for as long as 4 weeks. Significant elevations in a similar pattern also were noted for the benign patients. However, the baseline and postoperative median values were lower. The clinical impact from increased blood levels of VEGF is uncertain. It is possible that the growth of residual tumor deposits may be stimulated early after surgery. These results warrant a larger study as well as endothelial cell in vitro assays to determine whether postoperative plasma stimulates proliferation and invasion.

Perioperative sargramostim (recombinant human GM-CSF) induces an increase in the level of soluble VEGFR1 in colon cancer patients undergoing minimally invasive surgery

INTRODUCTION

Experimentally, laparotomy is associated with increased tumor growth. In humans, abdominal surgery is associated with immunosuppression and elevated plasma VEGF levels that might stimulate tumor growth early after surgery. Avoidance of these surgery-related changes and their consequences may be advantageous. Granulocyte-macrophage colony stimulating factor (GMCSF) is a non-specific immune system up-regulator that has also been associated, experimentally, with increased release of soluble VEGF Receptor 1 (sVEGFR1) which is an endogenous inhibitor of VEGF. This study's purpose was to determine the impact of perioperatively administered recombinant human GMCSF (rhu-GMCSF) on both immune function and plasma sVEGFR1 levels in colorectal cancer patients.

METHODS

This randomized placebo-controlled study included 36 colorectal cancer patients who underwent minimally invasive resection (17 GMCSF, 19 Placebo). Patients received 7 subcutaneous injections of either rhu-GMCSF, 125 microg/m2, or saline on preoperative days 3, 2 and 1 and on postoperative days (POD) 1, 2, 3 and 4. A number of immune parameters were followed and plasma levels of soluble VEGF Receptor 1 (sVEGFR1) and VEGF were determined.

RESULTS

The total WBC, neutrophil, eosinophil, and monocyte counts were significantly higher after surgery in the GMCSF group; no differences were noted for the other immune parameters. In the GMCSF group, median plasma sVEGFR1 levels were significantly elevated on POD 1 (188.1 pg/ml), and on POD 5 (142.8 pg/ml) when compared to pre-GMCSF levels (0 pg/ml) (p-value<0.05 for all comparisons). In the placebo group, the POD5 median sVEGFR1 level (116.3 pg/ml) was elevated and of borderline significance (p=0.05) vs the pre-treatment result (0 pg/ml). Of note, both groups had significantly elevated median plasma VEGF levels on POD 5 (Control 435.7 pg/ml; GMCSF 385.3 pg/ml) when compared to their preoperative results (Control 183.3 pg/ml, p=0.0013; GMCSF 171.5 pg/ml, p=0.0055).

CONCLUSIONS

Perioperative GMCSF was not associated with an immune function benefit in this study, however, such treatment leads to increased plasma sVEGFR1 levels. Colorectal resection, with or without GMCSF, was also associated with increased VEGF levels postoperatively. Increased plasma levels of sVEGFR1 after surgery might limit the pro-angiogenic tumor stimulatory effects of VEGF. Further study of GMCSF's impact on angiogenesis appears warranted.

Insulin-like growth factor binding protein-3 inhibits colitis-induced carcinogenesis

PURPOSE

Chronic inflammation in the setting of inflammatory bowel disease is thought to result in altered epithelial cell growth regulation and ultimately carcinogenesis. This loss in cell growth regulation may be partially caused by a decrease in circulating intact insulin-like growth factor binding protein-3 (IFGB-3) as a result of chronic inflammation. This study evaluates the effect of IFGB-3 on carcinogenesis in the setting of colitis.

METHODS

A previously described animal model for colitis-induced carcinogenesis was used. Colitis was induced in both wild-type and IFGB-3 transgenic CD1 mice with a one-week oral exposure to dextran sodium sulfate (2 percent in drinking water). All mice received a single intraperitoneal administration (10 mg/kg body weight) of a genotoxic colonic carcinogen, azoxymethane. At Week 20, the animals were killed and their colons were excised. The colons were examined by a pathologist under blinded conditions. Criteria assessed included the severity of colitis, number of aberrant crypt foci per mouse colon, incidence of colonic adenomas, and mean size of colonic adenomas.

RESULTS

A total of 20 mice (10 in each group) were included in the study. The severity of colitis was not significantly different between the two groups (mean colitis score wild-type = 13.2; IFGB-3 transgenic = 11; P = not significant). The average number of aberrant crypt foci per colon was significantly lower in the IFGB-3 transgenic mice compared with the wild-type mice (1.5 +/- 1.4 vs. 4.5 +/- 2.7, respectively; P < 0.0001). The number of adenomas per colon was significantly lower in IFGB-3 transgenic group (1.2 +/- 1.8) compared with the wild-type mice (3.7 +/- 2.7; P = 0.005). In addition the average size of adenomas was significantly smaller in IFGB-3 transgenic mice (1.4 +/- 1.3 mm) compared with the wild-type mice (2.6 +/- 2 mm; P = 0.013).

CONCLUSIONS

IFGB-3 significantly reduces the development of colonic tumors and precursor lesions in the setting of induced murine colitis. It is possible that the loss of IFGB-3 as a result of chronic inflammation may be associated with an increased rate of carcinogenesis in the inflammatory bowel disease setting. Although further studies are necessary, in theory, inhibiting the depletion of IFGB-3 or replacement of IFGB-3 may serve as a novel treatment strategy to prevent the development of colitis-induced carcinogenesis.

Major surgical trauma induces proteolysis of insulin-like growth factor binding protein-3 in transgenic mice and is associated with a rapid increase in circulating levels of matrix metalloproteinase-9

BACKGROUND

The authors previously demonstrated a significant decrease in plasma levels of intact insulin-like growth factor binding protein-3 (IGFBP-3) after major open but not after laparoscopic-assisted surgery in humans. They postulated that this decrease may have an effect on postoperative tumor growth. It also has been shown that plasma levels of matrix metalloproteinase-9 (MMP-9), a protease capable of degrading IGFBP-3, are transiently increased after open colectomy in humans. The authors aimed to develop an animal model that would allow further study of the effect that surgical trauma has on plasma levels IGFBP-3 and MMP-9. In addition, they set out to assess the concentration of MMP-9 in circulating monocytes before and after surgery.

METHODS

The 30 mice included in this study were divided into three groups: sham laparotomy, carbon dioxide (CO2) pneumoperitoneum, and anesthesia control. All mice were IGFBP-3 transgenics (overexpressing human IGFBP-3) on a CD1 background. The mice were anesthetized using ketamine and xylazine. Blood was drawn retroorbitally 48 h before the procedure. The duration of the procedure was 30 min. The animals were killed 24 h postoperatively and blood was drawn. Intact IGFBP-3 levels were measured using a combination of Western blot analysis and enzyme-linked immunoassay (ELISA) at the two time points: before and after the operation. Plasma and peripheral blood mononuclear cell levels of MMP-9 were measured at each time point using zymography. Mononuclear cell lysates were used to determine intracellular MMP-9 levels.

RESULTS

Plasma levels of intact IGFBP-3 were significantly lower than preoperative levels after sham laparotomy. A mean decrease of 76.6% was noted (p < 0.05). Zymography demonstrated significantly higher plasma MMP-9-related proteolytic activity than observed preoperatively after sham laparotomy (78.5 vs 42.3 Relative Units [RU]; p < 0.05). In the pneumoperitoneum group, no significant decrease was found between the pre- and postoperative levels of intact IGFBP-3. A nonsignificant increase in MMP-9 was noted after CO2 pneumoperitoneum (38 RU preoperatively vs. 46.4 RU postoperatively; p > 0.05). The anesthesia control group did not demonstrate a significant change in either circulating intact IGFBP-3 levels or MMP-9 levels. Mononuclear intracellular levels of MMP-9 were significantly lower after laparotomy than the preoperative levels (3 vs 37 RU). The postprocedure intracellular levels of MMP-9 were not significantly decreased in the pneumoperitoneum or anesthesia control group.

CONCLUSION

Plasma levels of intact IGFBP-3, a cell growth regulating factor, were found to be decreased significantly after laparotomy. This decrease was not seen after pneumoperitoneum. Depletion of intact IGFBP-3 after laparotomy correlated with a rapid release of MMP-9 from mononuclear cells and an increase in circulating plasma MMP-9 levels. Matrix metalloproteinase-9 may play an important role in IGFBP-3 proteolysis after surgical trauma. Furthermore, circulating mononuclear cells are one source of MMP-9 after surgery. Finally, the model used reproduces events in humans after surgery, and thus should permit further study on the mechanism of IGFBP-3 proteolysis after surgical trauma.

Converted laparoscopic colectomy: what are the consequences?

BACKGROUND

The safety and benefits of laparoscopic colon resection are well documented. However, few reports have addressed the safety and comparative outcome of laparoscopic colon operations that necessitated conversion.

METHODS

All consecutive laparoscopic colon resections performed by a single surgeon from July 1996 to October 2003 were assessed. Data obtained from a prospective computerized database included demographics, diagnosis, reason and time to conversion, length of stay, morbidity, and mortality. Additionally, all laparoscopic-converted colectomies were then matched with open colectomies by diagnosis and severity of disease and analyzed with respect to morbidity, mortality, and clinical outcome.

RESULTS

A total of 143 laparoscopic colon resections were analyzed, 78 of which were left colon resections and 65 were right colon resections. The overall conversion rate was 19.6% (28 patients). The disease entities of the 28 converted patients were diverticulitis (16), polyps (four), Crohn's disease (three), metastatic cancer (three), and others (two). Conversion was higher in the left-sided (24 patients, 30.8%) versus right-sided (four patients, 6.1%) procedures. There were no differences regarding age, gender, and comorbidities among the laparoscopic, open, and converted groups; the median follow-up was 39 months. The median length of stay was 6, 8, and 12 days for the laparoscopic, open, and converted groups, respectively. Right-sided conversions were due to the size of the inflammatory mass in three patients and intraoperative bleeding in one patient. Left-sided conversions were due to the inflammatory process extending beyond the sigmoid colon in 12 patients, adhesions in five, obesity in four, pericolonic abscess in two, and fixed mass in one patient. Postoperative morbidity was significantly higher for laparoscopic procedures that were converted to open procedures more than 30 min into the operation. Preoperative predictors of conversion were extent of inflammatory process beyond the sigmoid colon and obesity, whereas intraoperative predictors were adhesions and bleeding.

CONCLUSIONS

Laparoscopic-converted colon resection is associated with significantly greater morbidity, particularly wound complications and greater length of hospital stay, compared to open or laparoscopic colectomies. Prompt conversion (<30 min) may reduce the overall morbidity associated with converted procedures. Furthermore, thoughtful patient selection may decrease the conversion rate and thereby prevent the inherent morbidity associated with converted procedures.

Altered plasma matrix metalloproteinase-9/tissue metalloproteinase-1 concentration during the early postoperative period in patients with colorectal cancer

BACKGROUND

The authors have previously demonstrated that insulin-like growth factor binding protein-3 (IGFBP-3) is depleted in plasma for 1 to 3 days after major open surgery (OS), but not after laparoscopic surgery (LS). After surgery, IGFP-3 cleavage occurs rapidly and is likely attributable to altered plasma proteolytic activity. This study aimed to assess plasma proteolysis after both open and closed colorectal resection and, if possible, to identify a protease/protease inhibitor system affected by surgery.

METHODS

Plasma from 88 patients with colorectal cancer (stages I-III) who underwent resection was obtained preoperatively (pre-OP) and on postoperative days (POD) 1 to 3. Plasma proteolytic activity was assessed via zymography. On the basis of the results, specific protease and protease inhibitor concentrations were next measured via enzyme-linked immunoassay (ELISA). Statistical analysis was performed using Wilcoxon's test.

RESULTS

Early after surgery, zymography showed a predominant band representing a 92-kDa gelatinase corresponding to a proform of matrix metalloproteinase-9 (MMP-9), a protease known to cleave IGFBP-3. In OS patients, the mean concentration of plasma MMP-9 was significantly higher on POD 1 than at pre-OP (p < 0.003). On POD 2 and 3, no differences were noted. In the LS group, the mean levels of MMP-9 before and after surgery were comparable. The levels of a natural MMP-9 inhibitor, tissue inhibitor of metalloproteinase-1 (TIMP-1), also were measured. In the OS group, the level of TIMP-1 was significantly higher on POD 1 (p < 0.0003) and POD 2 (p < 0.01) and 3 (p < 0.01) than at pre-OP. In the LS group, a smaller but significant increase in TIMP-1 levels was found between the pre-OP sample and the POD 1 (p < 0.01) and POD 2 (p < 0.01) samples. No difference was noted on POD 3 (p = 0.1).

CONCLUSIONS

Open surgery, but not laparoscopic surgery, is accompanied by a short-lived significant increase in MMP-9 levels, which likely accounts for the decrease in IGFBP-3 levels observed after OS. The transitory nature of MMP-9 imbalance may be attributable to the increase in TIMP-1 levels postoperatively.