Discussion
This Australian study of 342 microvascular free flap head and neck reconstructions investigated whether operative time is associated with post-operative complications and length of stay. In this patient cohort, lengthier operative times, particularly those longer than 9 hours duration, were associated with adverse outcomes including surgical complications, any complication, return to theatre and length of stay after adjusting for the effect of potential confounders such as comorbidity, patient age, tumor site, reconstruction type, and use of tracheostomy.
In our analysis, the adjusted effect of one hour of additional operative time was a 13% increase in the odds of developing a surgical complication (p=0.002), a 20% increase in the odds of returning to theatre (p<0.001), and one extra day of hospital stay (p<0.001). This equated to more than double the odds of a surgical complication for operations > 9 hours duration compared to ≤ 9 hours duration (p=0.005). These findings are not unique and are supported by several studies from other countries showing that prolonged operative time is an independent risk factor for post-operative complications.8,10,11 In fact, Serleti et al. identified a similar cut-off (≥10 hours) as being associated with complications such as flap failure, thromboembolism, bleeding, and hematoma.13 Eskander et al. found that each hour was associated with a 21% increase in the odds of a complication and a 24% increase in the odds of wound-healing problems.1However, other studies have shown dramatic variations in the critical time (cut-off) associated with adverse outcomes, ranging from 480 minutes to 920 minutes, highlighting the importance of institutional factors and the need for individual units to publish their own results.8,11
Unfortunately, our cohort was not large enough to analyze whether operative time predicted free flap failure. However, two national database studies have shown that this is the case, including Ishimaru et al.14 who analyzed 2846 patients from the Japanese national database and Sanati-Mehrizy et al.15 who analyzed 2013 patients from the American College National Surgical Quality Improvement Program database. The association between operative time and surgical morbidity is robust across multiple specialties. This is well demonstrated in a meta-analysis by Cheng et al. that found a statistical association in 80% of the 66 studies they included.16 Whilst the critical time-threshold varied between studies (e.g. 2, 4, 6 hours), it is remarkable that pooled adjusted risk of complications consistently doubled in all specialties, and each additional hour was associated with a 21% increase in complication rate (p < 0.001), similar to our results.
Determining the true independent effect of operative time is almost impossible because it is dependent on multiple confounding factors, including case complexity, surgeon experience and technique, institutional factors such as operating room staff and protocols, prior treatment (surgery or radiotherapy), and intraoperative problems, which inevitably lengthen the operation and increase the likelihood of postoperative complications. Some of these factors cannot be modulated, however, head and neck microvascular reconstructions are typically long operations, where many components can be made more efficient. These include anaesthetic preparation time, surgical planning, equipment preparation, team communication and multi-tasking, and concurrent flap harvest with tumour ablation, as well as innovations in harvest and anastomotic techniques, such as advanced sealing technology and venous coupler use.8,11,12,17 In our institution we routinely performing concurrent flap harvest where possible and is likely to have the greatest impact on the duration of surgery. Several approaches make this option viable in a greater proportion of cases. Firstly, the ALT is the most common soft tissue flap in this cohort. This flap has the advantage of being able to harvest a larger portion of tissue than that anticipated to be necessary without substantially increasing the donor site morbidity. Secondly, we frequently use virtual surgical planning for osseous flap reconstruction. This allows the defect to be defined prior to surgery and has been shown in several studies to reduce operative time.18,19 Finally, we perform operative tracheostomies at the end of the operation and avoid elective tracheostomy use when it is felt safe to do so by both the ablative and reconstructive surgeon. Tracheostomy not only lengthens the procedure but was associated with double the odds of a complication in this series (Figure 1). Chaukar et al. found that both tracheostomy and longer operative time were independent predictors of wound complications after head and neck surgery, suggesting contamination of the wound with tracheal secretions as a likely mechanism.20
A prolonged operative time often indicates longer free flap ischemic time and tissue damage due to anoxic injuries. Moreover, the risk of reperfusion injury and the incidence of no flow phenomenon increase with longer ischemic time.21,22 Alternatively, increased operative time could be a proxy for procedure complexity, associated anaesthesia risks, or intraoperative complications.23Designing a study to control for these metrics is challenging as there are no validated measures of procedure complexity in head and neck oncology or reconstruction.