Colorectal cancer surgery in portal hypertensive patients: does adjuvant oxaliplatin affect prognosis?

BACKGROUND

Oxaliplatin is used in adjuvant treatment of colorectal cancer and is associated with sinusoidal obstruction syndrome. Few data are available on its effects in patients in whom portal hypertension was diagnosed before cancer treatment.

OBJECTIVE

Our aim was to investigate short- and long-term outcomes of surgery for colorectal cancer in patients with portal hypertension with or without cirrhosis, particularly regarding effects of adjuvant chemotherapy with oxaliplatin.

DESIGN AND SETTING

This was a prospective cohort study performed at an academic medical center.

PATIENTS

Patients with stage II or III colorectal cancer and portal hypertension who underwent curative resection were included.

INTERVENTION

All patients received adjuvant chemotherapy with oxaliplatin (FOLFOX 4) or 5-fluorouracil and leucovorin.

MAIN OUTCOME MEASURES

Potential predictive laboratory and clinical variables and postoperative (30-day) and long-term morbidity and mortality were recorded.

RESULTS

Of 63 patients enrolled, 23 (37%) had a total of 82 postoperative complications; 5 patients (8%) died within 30 days postoperatively. Univariate analysis showed that severe portal hypertension, preoperative Child class B, low albumin, the presence of ascites, preoperative upper GI tract bleeding, and high intraoperative blood loss were linked to postoperative morbidity. Presence of postoperative infection (p = 0.004), presence of preoperative ascites (p = 0.01), high intraoperative blood loss (p = 0.02), and preoperative upper GI tract bleeding (p = 0.03) were significantly related to mortality. Of 58 patients receiving adjuvant chemotherapy, 20 received the oxaliplatin regimen and 38 received 5-fluorouracil/leucovorin without oxaliplatin. The median length of follow-up was 26 (range, 6-36) months. Kaplan-Meier analyses showed that patients who received oxaliplatin had higher cumulative incidences of newly developed esophageal varices (p = 0.002), GI tract bleeding (p = 0.02), and newly formed ascites (p = 0.03). Death occurred in 8 of 20 patients (40%) in the oxaliplatin group and in 5 of 38 patients (13%) in the 5-fluorouracil group. Kaplan-Meier estimates of mean survival time were 34.4 months (95% CI, 32.4-36.5) in the 5-fluorouracil/leucovorin group vs 29.9 months (95% CI, 26-33.7) in the oxaliplatin group, and patients receiving oxaliplatin had a significantly higher relative risk of death (HR = 2.98; 95% CI, 1.03-8.65). Cancer-specific mortality was not related to treatment type.

LIMITATIONS

The study was limited by the relatively small sample size and lack of randomization, which may have led to selection bias in treatment regimens.

CONCLUSIONS

Colorectal cancer surgery can be done safely in portal hypertensive patients with good hepatic function; however, higher mortality is expected in patients with compromised hepatic function reserve. Compared with adjuvant chemotherapy without oxaliplatin, oxaliplatin-based chemotherapy does not significantly reduce cancer-specific mortality and may increase overall morbidity and mortality. Therefore, oxaliplatin-based chemotherapy should be used with caution in patients who have portal hypertension, even in those with good liver function.