Cervical variations of the phrenic nerve

OBJECTIVES/HYPOTHESIS: Selective reinnervation of the posterior cricoarytenoid muscle with a single phrenic nerve rootlet has been shown to restore physiologic motion in animal models. However, clinical translation of this work is challenged by the limited knowledge of the cervical anatomy of the phrenic nerve. STUDY DESIGN: Prospective collaborative study. METHODS: Dissection of 111 cadaveric necks (88 embalmed and 23 unembalmed) from 56 cadavers. RESULTS: The mean (standard deviation) lengths of unembalmed cadaver C3, C4, and C5 nerve rootlets were 3.9 (2.4), 3.6 (2.6), and 0.5 (0.8) cm, respectively. Embalmed cadavers had shorter C3 and C4 phrenic nerve rootlet lengths than unembalmed cadavers (P = .02 and P = .03, respectively). There was no difference in mean nerve rootlet length based on sex, body height or weight, or side of dissection. A total of eight unique phrenic nerve rootlet patterns were identified. The most common pattern consisted of phrenic with single C3 and C4 rootlets with an immeasurable C5 rootlet, which was present in 30 of 111 (26%) of the necks. The classic three branching pattern of single C3, C4, and C5 rootlets was found in 25 of 111 (22%) of the necks. Six of 111 (5%) of the dissections displayed accessory phrenic nerves arising from the C3, C4, or C5 anterior rami. A χ(2) analysis showed no difference between side or sex and frequency of pattern. CONCLUSIONS: The present study demonstrates the wide variability within the cervical anatomy of the phrenic nerve.