The Harmonic Scalpel versus Conventional Hemostasis for Neck Dissection: A Meta-Analysis of the Randomized Controlled Trials

Abstract

Objective: Neck dissection is the most definitive and effective treatment for head and neck cancer. This systematic review aims to compare the efficacy and surgical outcomes of neck dissection between the harmonic scalpel and conventional surgical techniques and conduct a quantitative meta-analysis of the randomized trials.

Methods: Randomized controlled trials (RCTs) were identified from the major electronic databases (MEDLINE, EMBASE and Cochrane Library) using the keywords ''harmonic scalpel'' and ''neck dissection,'' and a quantitative meta-analysis was conducted. The operative time and intraoperative bleeding were the primary outcome measures, and other parameters assessed included the drainage fluid volume and length of hospital stay.

Results: Seven trials that met the inclusion criteria included 406 neck dissection cases (201 in the harmonic scalpel group). Compared with conventional surgical techniques, the HS group had an operative time that was significantly reduced by 29.3 minutes [mean difference: -29.29; 95% CI = (-44.26, -14.32); P=0.0001], a reduction in intraoperative bleeding by 141.1 milliliters [mean difference: -141.13; 95% CI = (-314.99, 32.73); P=0.11], and a reduction in drainage fluid volume by 64.9 milliliters [mean difference: -64.86; 95% CI = (-110.40, -19.32); P=0.005] , but it is not significant after removal of studies driving heterogeneity. There was no significant difference in the length of the hospital stay [mean difference: -0.21; 95% CI = (-0.48, 0.07); P=0.14].

Conclusion: This systematic review showed that using the harmonic scalpel for neck dissection significantly reduces the operative time and drainage fluid volume and that it is not associated with an increased length of hospital stay or perioperative complications. Therefore, the harmonic scalpel method is safe and effective for neck dissection. However, the statistical heterogeneity was high. Further studies are required to substantiate our findings.

Fig 1. Flow diagram of literature search.

Fig 2. Risk of bias summary.

Two studies were identified as being of a higher design quality. Three studies were identified as being of a lower design quality. The quality of the remaining two studies was moderate. (“+”means the bias is low risk, “?” means the bias is unclear).

Fig 3. Forest plot for operative time.

The operative time with the harmonic scalpel was significantly shorter than that with conventional hemostasis, but the statistical heterogeneity was unacceptably large (I² = 92%).

Fig 4. Forest plot for the operative time (sensitivity analysis).

Heterogeneity became acceptable (I² = 48%) and the effect measure remained significant (P<0.00001).

Fig 5. Forest plot for blood loss.

The intraoperative blood loss with the harmonic scalpel was shorter than that with conventional hemostasis, but not significantly, and the statistical heterogeneity was unacceptably large (I² = 100%).

Fig 6. Forest plot for blood loss(sensitivity analysis).

Heterogeneity became acceptable (I² = 0%) and the effect measure was significant(P<0.00001).

Fig 7. Forest plot for the amount of drainage.

Total drainage fluid volume with the harmonic scalpel was significantly shorter than that with conventional hemostasis but the statistical heterogeneity was unacceptably large (I² = 97%).

Fig 8. Forest plot for the amount of drainage (sensitivity analysis).

Heterogeneity became acceptable (I² = 0%) but the effect measure was not significant (P = 0.58).

Fig 9. Forest plot for the hospital stay.

Heterogeneity was acceptable (I² = 0%). However, the difference between the two groups was not significant (P = 0.14).