Dual VTE Prophylaxis Warranted for Colectomy Patients

MADISON, WIS. – Only combined pharmacologic and mechanical prophylaxis was protective against venous thromboembolism following colectomy in a large prospective analysis of 3,464 patients.

Major bleeding events were significantly lower at 0.3% with combination prophylaxis, compared with 1.3% without combination prophylaxis (P = .005).

Patrice Wendling/IMNG Medical Media

There was no difference in major bleeding events with or without pharmacologic prophylaxis alone (0.6% vs. 0.7%; P = .45).

Counterintuitively, patients who received only sequential compression devices as prophylaxis had more major bleeding events than those who did not receive devices (1.6% vs. 0.4%; P = .009), vascular surgeon Dr. Peter Henke said at the annual meeting of the Central Surgical Association.

"All [colectomy patients] should have combined mechanical and pharmacologic VTE prophylaxis, unless their bleeding risk is deemed too high by surgeon judgment," he said.

Investigators at 15 hospitals prospectively collected data from 2008 to 2009 on 3,464 open and laparoscopic colectomy patients. Their mean age was 65 years, and 53% were female. Roughly 74% of patients were on combination prophylaxis during both years.

The VTE incidence was low overall at 2.2%, but as high as other high-risk surgical procedures such as total knee or hip replacement, said Dr. Henke, with the University of Michigan Health System in Ann Arbor.

VTEs occurred slightly more often with open left colectomy than with open right colectomy (3.1% vs. 2.3%). VTE rates for laparoscopic left and right colectomies were significantly lower (1.7% and 0.5%; P = .011), while rates were double for emergent vs. nonemergent cases (4% vs. 2%; P = .017).

In a univariate analysis of 2,263 patients with full data available, several significant risk factors for postoperative VTE emerged, including no combination prophylaxis; older age (65 vs. 62 years); higher body mass index (31 vs. 28 kg/m²); preoperative angina, anemia, and weight loss; surgical site infections; and postoperative infectious complications.

Notably, operative time, presence of malignancy, anastomotic leak, transfusion, and urinary tract infection were not significantly associated with VTE.

In a full multivariate model, postoperative infectious complications increased the risk for a postoperative VTE (odds ratio, 3.6; P less than .001), Dr. Henke said. This finding has also been observed in other studies.

Risk was also significantly higher with contaminated wounds (OR, 3.4; P = .001), postoperative surgical site infection (OR, 2.5; P = .01), anemia (OR, 2.4; P = .01), each additional year of age (OR, 1.05, P = .04), and increased BMI (OR, 1.03, P = .01). VTE risk increased with open as compared to laparoscopic surgery, but not with covariate adjustment, he said.

Protective factors included no history of preoperative angina (OR, 0.18; P = .01) or weight loss (OR, 0.33; P = .02).

The only really modifiable protective factor, however, was combined mechanical and pharmacologic prophylaxis (OR, 0.48; P = .02; area under the curve, 0.81), Dr. Henke said.

The investigators also performed a propensity-matched analysis of 1,006 patients for anticoagulant type and VTE in which unfractionated heparin was equivalent to low-molecular-weight heparin (1.6% vs. 1.4%; P = .5).

Limitations of the study were the determination of VTE on clinical grounds and not prospectively by protocol, the inability to assess for hypercoagulable states or prior history of VTE – a strong risk factor for recurrent VTE – and follow-up of only 30 days, while about 20% of VTEs occur outside this window. The findings also may not be reflective of rectal and small bowel surgical VTE risk, he said.

Dr. Susan Galandiuk, an invited discussant, asked why the analysis was limited to patients undergoing segmental colectomy, given that VTE rates and operating times are higher among patients undergoing rectal surgery. She also questioned why the group combined deep vein thrombosis (DVT) and pulmonary embolism into a VTE statistic. Dr. Galandiuk, with the University of Louisville (Ky.), noted that extensive work performed at her university has shown that while DVT is far more common, it is never really lethal, and although rare, pulmonary embolism is occasionally lethal.

Dr. Henke said the analysis focused on only four CPT codes to get as homogeneous a sample as possible, but agreed that rectal cases would be at even higher VTE risk and thus warrant combined prophylaxis. He also agreed that DVTs are not directly fatal, whereas pulmonary embolisms are, adding that interestingly, surgical patients tend to have less risk of fatal pulmonary embolism than do medically ill patients.

Dr. M. Ashraf Mansour, another invited discussant, who is interim chair of surgery at Michigan State University, Grand Rapids, asked how Dr. Henke would manage patients with a prophylaxis allergy or heparin-induced thrombocytopenia (HIT) and whether he would still recommend dual prophylaxis for patients with low DVT risk on the Caprini Risk Assessment Model.

Dr. Henke said parinox may be safe for patients with a history of HIT unless they are at really high risk, but that filters are overused and controversial. He noted that the same hospital consortium is trying to validate the Caprini score among 10,000 patients, with 90-day data now available on 8,500 patients.

"The interesting thing about this study that struck me initially as different is that, regardless of the other preoperative risk factors, dual prophylaxis was independently protective regardless of risk factors," he added. "So even with a low-risk Caprini score, they should probably have dual prophylaxis just because the overall [VTE] rate is quite high compared to many other surgical procedures."

Dr. Henke, Dr. Galandiuk and Dr. Mansour reported no relevant conflicts of interest.

MADISON, WIS. – Only combined pharmacologic and mechanical prophylaxis was protective against venous thromboembolism following colectomy in a large prospective analysis of 3,464 patients.

Major bleeding events were significantly lower at 0.3% with combination prophylaxis, compared with 1.3% without combination prophylaxis (P = .005).

Patrice Wendling/IMNG Medical Media

There was no difference in major bleeding events with or without pharmacologic prophylaxis alone (0.6% vs. 0.7%; P = .45).

Counterintuitively, patients who received only sequential compression devices as prophylaxis had more major bleeding events than those who did not receive devices (1.6% vs. 0.4%; P = .009), vascular surgeon Dr. Peter Henke said at the annual meeting of the Central Surgical Association.

"All [colectomy patients] should have combined mechanical and pharmacologic VTE prophylaxis, unless their bleeding risk is deemed too high by surgeon judgment," he said.

Investigators at 15 hospitals prospectively collected data from 2008 to 2009 on 3,464 open and laparoscopic colectomy patients. Their mean age was 65 years, and 53% were female. Roughly 74% of patients were on combination prophylaxis during both years.

The VTE incidence was low overall at 2.2%, but as high as other high-risk surgical procedures such as total knee or hip replacement, said Dr. Henke, with the University of Michigan Health System in Ann Arbor.

VTEs occurred slightly more often with open left colectomy than with open right colectomy (3.1% vs. 2.3%). VTE rates for laparoscopic left and right colectomies were significantly lower (1.7% and 0.5%; P = .011), while rates were double for emergent vs. nonemergent cases (4% vs. 2%; P = .017).

In a univariate analysis of 2,263 patients with full data available, several significant risk factors for postoperative VTE emerged, including no combination prophylaxis; older age (65 vs. 62 years); higher body mass index (31 vs. 28 kg/m²); preoperative angina, anemia, and weight loss; surgical site infections; and postoperative infectious complications.

Notably, operative time, presence of malignancy, anastomotic leak, transfusion, and urinary tract infection were not significantly associated with VTE.

In a full multivariate model, postoperative infectious complications increased the risk for a postoperative VTE (odds ratio, 3.6; P less than .001), Dr. Henke said. This finding has also been observed in other studies.

Risk was also significantly higher with contaminated wounds (OR, 3.4; P = .001), postoperative surgical site infection (OR, 2.5; P = .01), anemia (OR, 2.4; P = .01), each additional year of age (OR, 1.05, P = .04), and increased BMI (OR, 1.03, P = .01). VTE risk increased with open as compared to laparoscopic surgery, but not with covariate adjustment, he said.

Protective factors included no history of preoperative angina (OR, 0.18; P = .01) or weight loss (OR, 0.33; P = .02).

The only really modifiable protective factor, however, was combined mechanical and pharmacologic prophylaxis (OR, 0.48; P = .02; area under the curve, 0.81), Dr. Henke said.

The investigators also performed a propensity-matched analysis of 1,006 patients for anticoagulant type and VTE in which unfractionated heparin was equivalent to low-molecular-weight heparin (1.6% vs. 1.4%; P = .5).

Limitations of the study were the determination of VTE on clinical grounds and not prospectively by protocol, the inability to assess for hypercoagulable states or prior history of VTE – a strong risk factor for recurrent VTE – and follow-up of only 30 days, while about 20% of VTEs occur outside this window. The findings also may not be reflective of rectal and small bowel surgical VTE risk, he said.

Dr. Susan Galandiuk, an invited discussant, asked why the analysis was limited to patients undergoing segmental colectomy, given that VTE rates and operating times are higher among patients undergoing rectal surgery. She also questioned why the group combined deep vein thrombosis (DVT) and pulmonary embolism into a VTE statistic. Dr. Galandiuk, with the University of Louisville (Ky.), noted that extensive work performed at her university has shown that while DVT is far more common, it is never really lethal, and although rare, pulmonary embolism is occasionally lethal.

Dr. Henke said the analysis focused on only four CPT codes to get as homogeneous a sample as possible, but agreed that rectal cases would be at even higher VTE risk and thus warrant combined prophylaxis. He also agreed that DVTs are not directly fatal, whereas pulmonary embolisms are, adding that interestingly, surgical patients tend to have less risk of fatal pulmonary embolism than do medically ill patients.

Dr. M. Ashraf Mansour, another invited discussant, who is interim chair of surgery at Michigan State University, Grand Rapids, asked how Dr. Henke would manage patients with a prophylaxis allergy or heparin-induced thrombocytopenia (HIT) and whether he would still recommend dual prophylaxis for patients with low DVT risk on the Caprini Risk Assessment Model.

Dr. Henke said parinox may be safe for patients with a history of HIT unless they are at really high risk, but that filters are overused and controversial. He noted that the same hospital consortium is trying to validate the Caprini score among 10,000 patients, with 90-day data now available on 8,500 patients.

"The interesting thing about this study that struck me initially as different is that, regardless of the other preoperative risk factors, dual prophylaxis was independently protective regardless of risk factors," he added. "So even with a low-risk Caprini score, they should probably have dual prophylaxis just because the overall [VTE] rate is quite high compared to many other surgical procedures."

Dr. Henke, Dr. Galandiuk and Dr. Mansour reported no relevant conflicts of interest.

MADISON, WIS. – Only combined pharmacologic and mechanical prophylaxis was protective against venous thromboembolism following colectomy in a large prospective analysis of 3,464 patients.

Major bleeding events were significantly lower at 0.3% with combination prophylaxis, compared with 1.3% without combination prophylaxis (P = .005).

Patrice Wendling/IMNG Medical Media

There was no difference in major bleeding events with or without pharmacologic prophylaxis alone (0.6% vs. 0.7%; P = .45).

Counterintuitively, patients who received only sequential compression devices as prophylaxis had more major bleeding events than those who did not receive devices (1.6% vs. 0.4%; P = .009), vascular surgeon Dr. Peter Henke said at the annual meeting of the Central Surgical Association.

"All [colectomy patients] should have combined mechanical and pharmacologic VTE prophylaxis, unless their bleeding risk is deemed too high by surgeon judgment," he said.

Investigators at 15 hospitals prospectively collected data from 2008 to 2009 on 3,464 open and laparoscopic colectomy patients. Their mean age was 65 years, and 53% were female. Roughly 74% of patients were on combination prophylaxis during both years.

The VTE incidence was low overall at 2.2%, but as high as other high-risk surgical procedures such as total knee or hip replacement, said Dr. Henke, with the University of Michigan Health System in Ann Arbor.

VTEs occurred slightly more often with open left colectomy than with open right colectomy (3.1% vs. 2.3%). VTE rates for laparoscopic left and right colectomies were significantly lower (1.7% and 0.5%; P = .011), while rates were double for emergent vs. nonemergent cases (4% vs. 2%; P = .017).

In a univariate analysis of 2,263 patients with full data available, several significant risk factors for postoperative VTE emerged, including no combination prophylaxis; older age (65 vs. 62 years); higher body mass index (31 vs. 28 kg/m²); preoperative angina, anemia, and weight loss; surgical site infections; and postoperative infectious complications.

Notably, operative time, presence of malignancy, anastomotic leak, transfusion, and urinary tract infection were not significantly associated with VTE.

In a full multivariate model, postoperative infectious complications increased the risk for a postoperative VTE (odds ratio, 3.6; P less than .001), Dr. Henke said. This finding has also been observed in other studies.

Risk was also significantly higher with contaminated wounds (OR, 3.4; P = .001), postoperative surgical site infection (OR, 2.5; P = .01), anemia (OR, 2.4; P = .01), each additional year of age (OR, 1.05, P = .04), and increased BMI (OR, 1.03, P = .01). VTE risk increased with open as compared to laparoscopic surgery, but not with covariate adjustment, he said.

Protective factors included no history of preoperative angina (OR, 0.18; P = .01) or weight loss (OR, 0.33; P = .02).

The only really modifiable protective factor, however, was combined mechanical and pharmacologic prophylaxis (OR, 0.48; P = .02; area under the curve, 0.81), Dr. Henke said.

The investigators also performed a propensity-matched analysis of 1,006 patients for anticoagulant type and VTE in which unfractionated heparin was equivalent to low-molecular-weight heparin (1.6% vs. 1.4%; P = .5).

Limitations of the study were the determination of VTE on clinical grounds and not prospectively by protocol, the inability to assess for hypercoagulable states or prior history of VTE – a strong risk factor for recurrent VTE – and follow-up of only 30 days, while about 20% of VTEs occur outside this window. The findings also may not be reflective of rectal and small bowel surgical VTE risk, he said.

Dr. Susan Galandiuk, an invited discussant, asked why the analysis was limited to patients undergoing segmental colectomy, given that VTE rates and operating times are higher among patients undergoing rectal surgery. She also questioned why the group combined deep vein thrombosis (DVT) and pulmonary embolism into a VTE statistic. Dr. Galandiuk, with the University of Louisville (Ky.), noted that extensive work performed at her university has shown that while DVT is far more common, it is never really lethal, and although rare, pulmonary embolism is occasionally lethal.

Dr. Henke said the analysis focused on only four CPT codes to get as homogeneous a sample as possible, but agreed that rectal cases would be at even higher VTE risk and thus warrant combined prophylaxis. He also agreed that DVTs are not directly fatal, whereas pulmonary embolisms are, adding that interestingly, surgical patients tend to have less risk of fatal pulmonary embolism than do medically ill patients.

Dr. M. Ashraf Mansour, another invited discussant, who is interim chair of surgery at Michigan State University, Grand Rapids, asked how Dr. Henke would manage patients with a prophylaxis allergy or heparin-induced thrombocytopenia (HIT) and whether he would still recommend dual prophylaxis for patients with low DVT risk on the Caprini Risk Assessment Model.

Dr. Henke said parinox may be safe for patients with a history of HIT unless they are at really high risk, but that filters are overused and controversial. He noted that the same hospital consortium is trying to validate the Caprini score among 10,000 patients, with 90-day data now available on 8,500 patients.

"The interesting thing about this study that struck me initially as different is that, regardless of the other preoperative risk factors, dual prophylaxis was independently protective regardless of risk factors," he added. "So even with a low-risk Caprini score, they should probably have dual prophylaxis just because the overall [VTE] rate is quite high compared to many other surgical procedures."

Dr. Henke, Dr. Galandiuk and Dr. Mansour reported no relevant conflicts of interest.