Growing rod fractures: risk factors and opportunities for prevention.

STUDY

DESIGN

Review of a prospectively collected growing rod database.

OBJECTIVE

To define risk factors for and characterize the nature of growing rod fractures.

SUMMARY OF BACKGROUND DATA: Rod fracture is a common complication of growing rod treatment. The project sought to analyze risk factors for rod breakage and develop preventive strategies.

METHODS

Records of 327 patients in a prospectively collected growing rod database were studied. Risk factors studied were studied as patient-related and rod-related. Multivariate analysis was performed.

RESULTS

Eighty-six rod fractures occurred in 49 patients (49 of 327, 15%). Sixteen patients had repeat fractures with eight patients having more than two fractures (maximum six). The most common fracture locations were above or below the tandem connectors (34 of 86) and near the thoracolumbar junction (35 of 86). Other locations were adjacent to anchors (12 of 86) and cross-links (2 of 86). Syndromic diagnoses had the highest rate of fracture; significantly greater than neuromuscular diagnoses (14% vs. 2%, P = 0.01). Patients who were ambulatory had a higher fracture rate (21% vs. 8.7%, P = 0.01). Single rods had a higher fracture rate than dual rods (36% vs. 11%, P < 0.001). Repeat fracture was also more common in patients with single rods (13% vs. 2%, P = 0.0002). In dual-rod constructs, the incidence of both rods breaking at the same time was 26% (7 of 27). Stainless steel rods had a higher fracture rate than titanium rods (29% vs. 18%, P = 0.02). The nonfracture group had larger diameter rods than the fracture group (P = 0.01). The fracture group had shorter tandem connectors than the nonfracture group (P < 0.001). Neither the size of preoperative scoliosis (P = 0.2) nor kyphosis (P = 0.4) was a risk factor for fracture. Length of instrumentation (P = 0.9), anchor type (P = 0.6), and pelvic fixation (P = 0.38) had no significant effect on fracture rates. Eight wound complications were reported, including three cases of skin breakdown at the rod fracture site.

CONCLUSION

Risk factors for rod fractures include prior fracture, single rods, stainless steel rods, small diameter rods, proximity to tandem connectors, short tandem connectors, and preoperative ambulation. Repeat fractures are common, especially with single rods. Rod replacement, with larger diameter rods if appropriate, may be a preferred strategy over connecting the broken rods as fractures signal fatigue of the rod.