Back to 2015 Annual Meeting Program
Complication Type and Severity Have an Adverse Effect on Long-Term Oncological Outcome Following Surgery for Colorectal Cancer
Stephen T. McSorley*, David G. Watt, Paul G. Horgan, Donald C. McMillan
Academic Unit of Surgery, University of Glasgow, Glasgow, United Kingdom
Background
Complications following surgery for colorectal cancer have been reported to have a negative impact on long-term outcomes. These complications have been defined by type, (infective or non-infective), and by severity using the Clavien Dindo scale. To date these two approaches have not been directly compared. The present study aimed to determine whether the type or severity of complication has the greatest impact on long-term outcomes following surgery for colorectal cancer.
Methods
Data was recorded prospectively for patients undergoing potentially curative surgery for colorectal cancer in one hospital between March 2008 and May 2013. Patients with metastatic disease or undergoing palliative surgery were excluded. Complications were retrospectively classified by type and by Clavien Dindo grade. Categorical data was analysed using the chi square test for independence. Survival data was analysed using Cox's proportional hazards model. P values <0.05 were considered significant.
Results
Of the 409 patients included the majority were male (55%), over 65 years old (67%), had elective surgery (92%), with colonic (66%) and node negative disease (65%). 155 patients (38%) suffered a complication, of which 105 were infective, and 31 were Clavien Dindo grade 3 or 4. Median follow up after surgery was 30 months with overall survival 82% and disease specific survival 89%. At univariate analysis infective complications were associated with poorer disease specific survival (HR 2.49, 95%CI 1.34-4.60, p=0.003) and poorer overall survival (HR 1.80, 95%CI 1.10-2.97, p=0.02) whereas non-infective complications were not (HR 1.74, 95%CI 0.74-4.08, p=0.19 and HR 1.52, 95%CI 0.78-2.96, p= 0.22 respectively). Severe complications (Clavien Dindo grade 3 to 4) were associated with poorer disease specific survival (HR 3.21, 95%CI 1.37-7.53, p=0.007) but not overall survival (HR 1.79, 95%CI 0.80-4.00, p=0.16). Strong agreement between the type and severity of complication meant only complication severity was taken forward into multivariate analysis with age, presentation (emergency or elective), TNM stage and adjuvant treatment (all p <0.01 at univariate analysis). At multivariate analysis complication severity (HR 1.56, 95%CI 1.05-2.34, p=0.03), TNM stage (HR 3.92, 95%CI 2.25-6.81, p<0.001) and adjuvant treatment (HR 0.35, 95% CI 0.16-0.78, p=0.01) all remained independently significant predictors of disease specific survival however complication severity was not significant in terms of overall survival (HR 1.19, 95%CI 0.84-1.69, p=0.32).
Conclusions
Although based on different definitions of complications, the type and severity were strongly associated and both predicted long-term outcomes, in particular disease specific survival. This would suggest a common underlying mechanism affecting long-term survival.
Table 1: Patient characteristics by complication severity
| Characteristic | All | Clavien Dindo complication grade |
| 0 | 1-2 | 3-4 | 5 | P |
| N(%) | 409(100) | 254(62) | 118(29) | 31(8) | 6(1) | NA |
| Age(<65/65-74/>74) | 136/155/118 | 88/99/67 | 38/45/35 | 10/9/12 | 0/2/4 | 0.275 |
| Gender (male/female) | 225/184 | 122/132 | 78/40 | 20/11 | 5/1 | 0.003 |
| Presentation (elective/emergency) | 377/32 | 239/15 | 108/10 | 26/5 | 4/2 | 0.021 |
| TNM stage (0/I/II/III) | 10/80/176/143 | 8/54/112/80 | 1/20/47/50 | 0/5/14/12 | 1/1/3/1 | 0.195 |
| Site (colon/rectum) | 268/141 | 173/81 | 73/45 | 20/11 | 2/4 | 0.239 |
| Neoadjuvant treatment (no/yes) | 330/65 | 205/40 | 94/19 | 26/5 | 5/1 | 0.999 |
| Adjuvant treatment (no/yes) | 282/121 | 181/73 | 75/43 | 26/5 | NA | 0.069 |
| Preop mGPS (0/1/2) | 294/43/72 | 185/26/43 | 84/11/23 | 20/6/5 | 5/0/1 | 0.698 |
| Preop NLR (<5/>5) | 339/70 | 205/49 | 104/14 | 27/4 | 3/3 | 0.044 |
| Approach (open/lap) | 297/111 | 177/77 | 93/24 | 23/8 | 4/2 | 0.258 |
| Infective complication (no/yes) | 50/105 | NA | 39/79 | 10/21 | 1/5 | 0.704 |
Clavien Dindo scale; 0 = no complication, 1-2 = complication requiring minor intervention, 3-4 = complication requiring significant intervention, 5 = death. mGPS preoperative modified Glasgow Prognostic score (0 = CRP<10mg/L, 1 = CRP≥10mg/L and albumin ≥35g/L, 2 = CRP≥10mg/L and albumin <35g/L). NLR preoperative neutrophil lymphocyte ratio. NA not applicable
Table 2: Effect of postoperative complication severity and type on disease specific and overall survival - univariate analysis
| Complication | DSS (%) | HR (95%CI) | P | | OS (%) | HR (95%CI) | P |
| Clavien-Dindo severity grade | | | | | | | | |
| 0 | 92 | 1.00 | ref | | 84 | 1.00 | ref |
| 1-2 | 86 | 1.82(0.96-3.43) | 0.065 | | 81 | 1.37(0.82-2.28) | 0.23 |
| 3-4 | 77 | 3.21(1.37-7.53) | 0.007 | | 77 | 1.79(0.80-4.00) | 0.16 |
| Infective vs. non-infective | | | | | | | | |
| No complication | 92 | 1.00 | ref | | 84 | 1.00 | ref |
| Infective complication | 82 | 2.51(1.36-4.64) | 0.003 | | 76 | 1.81(1.10-2.99) | 0.02 |
| Non-infective complication | 86 | 1.74(0.74-4.08) | 0.2 | | 78 | 1.52(0.78-2.96) | 0.22 |
OS overall survival, DSS disease specific survival, HR hazard ratio, CI confidence interval. Clavien Dindo complication grades; 0 = no complication, 1-2 = complication requiring minor intervention, 3-4 = complication requiring significant intervention.
Back to 2015 Annual Meeting Program
