Gregory S Kazarian, Jung K Mok, Mihir Dekhne, Stephane Owusu-Sarpong, Takashi Hirase, James Farmer, Russel C Huang, Han Jo Kim, Sravisht Iyer, James E Dowdell, Todd J Albert, Ellen M Soffin, James D Beckman, Oliver P F Panzer, Douglas S Wetmore, Sheeraz A Qureshi, Kyle W Morse
Cervical hematoma after ACDF/CDR occurred in 0.41% of cases and was independently predicted by male sex and ≥4 surgical levels. ENT approach did not affect risk after matching. Penrose drains were associated with higher hematoma rates than closed-suction or no drain. These findings inform perioperative risk stratification and surgical decision-making.
BACKGROUND CONTEXT: Postoperative cervical hematoma requiring surgical evacuation after anterior cervical discectomy and fusion (ACDF) and cervical disc replacement (CDR) is rare but life-threatening. Risk factor data are limited and inconsistent, largely derived from heterogeneous administrative databases.
PURPOSE: To characterize the incidence and independent risk factors for postoperative cervical hematoma in a large single-institution consecutive series, and to evaluate the roles of ENT involvement and drain type.
DESIGN: Retrospective cohort study.
PATIENT SAMPLE: 6,084 consecutive ACDF/CDR patients at a single academic institution; 25 confirmed hematoma cases identified by manual chart review.
OUTCOME MEASURES: Cervical hematoma requiring surgical evacuation.
METHODS: Univariate and multivariable logistic regression; 1:1 propensity score matching without replacement (caliper 0.2xSD) for ENT involvement and drain type. Post hoc power was assessed for key comparisons.
RESULTS: Incidence was 0.41% (25/6,084). On univariate analysis, male sex (80.0% vs 56.6%, p=0.024), clotting disorder (OR 3.78, p=0.017), postoperative antihypertensive use (OR 2.36, p=0.042), and ≥4 surgical levels (OR 4.19, p=0.022) were associated with increased risk. On multivariable analysis, male sex (aOR 2.99, p=0.029) and ≥4 surgical levels (aOR 3.20, p=0.037) were independent predictors. Surgical approach by ENT showed no association after propensity matching (OR 1.00, p=1.000). Penrose drains were associated with higher rates than closed-suction drains (OR 6.08, p=0.001) or no drain (OR 9.15, p<0.001).
CONCLUSIONS: Cervical hematoma after ACDF/CDR occurred in 0.41% of cases and was independently predicted by male sex and ≥4 surgical levels. ENT approach did not affect risk after matching. Penrose drains were associated with higher hematoma rates than closed-suction or no drain. These findings inform perioperative risk stratification and surgical decision-making.