Perry and Nickel1 introduced the halo in 1959 to address clinical situations requiring the need for rigid cervical spine immobilization. The halo provides the most rigid immobilization of all commercially available cervical orthoses.2 Halo orthoses have improved considerably with regard to component selection and availability, vest fit, and advanced application techniques.3 Since the introduction of the halo-vest, several studies documented complications associated with halo immobilization in both children and adults.4,7 Skull pin loosening is the most common complication occurring with use of the halo vest.5,8 Efforts to reduce the incidence of pin related complications have increased the importance of halo vest fit. Mirza et al.9 studied the effects of halo vest fit and its influence on overall halo stability, determining that fit of the vest on the torso, interface friction, rigidity of the vest material and arrangement and stiffness of the superstructure are all important. Stability varied with vest fit and continual adjustment and monitoring of the fit of the vest were crucial in maintaining efficacy of the device. Along with improved vest fit, the manner and techniques in which the halo is applied are also important in the overall halo application. To improve overall pin stability, a consistent protocol for component application and pin insertion is required. This protocol should not vary the method and order of pin insertion, amount of torque, number of pins used, pin insertion angle, location on pin placement, and the use of stab incisions. Stability at the pin-bone interface is thought to translate into a more favorable halo construct and thereby decrease the rate of commonly occurring complications.10,11 To effectively test the incidence of pin loosening as related to one pin type or another in a patient population, a standard application protocol was used to attempt to eliminate confounding variables. A consistent halo application technique is critical. This must include strict methodology of crown and pin placement as well as vest placement. With this in place, the incidence of pin loosening between two types of skull pins was investigated.
Two different types of halo pins (Figure 1) were tested in a prospective manner using data collection techniques of clinical evaluations and subjective questionnaires. The standard halo pin (DePuy AcroMed, Raynham, MA; Figure 1B) is designed with a sharp conical tip that flares into a rounded shoulder. This standard pin is hypothesized to function with the initial forces of pin insertion, producing a wedge effect that stabilizes the pin in the bone surface. The compressive load can exceed the compressive yield strength of the bone and may lead to subsequent bony necrosis around the pin and clinical loosening of the halo pin.12 The new trochar-style HIFix™ pin (DePuy, AcroMed; Figure 1A) is designed with a cylindrical tip and a cutting edge. The theory behind the design of the HIFix pin is that the cylindrical tip of the pin will cut into the adjacent bone during pin insertion. The cutting tip therefore minimizes damage to the adjacent bone and does not rely on high axial forces to maintain fixation. The compressive forces will therefore be lower, and less damage will occur to the bone, therefore increasing the stability and fixation.13
MATERIALS AND METHODS
Study design consisted of monitoring the incidence of pin loosening among 40 patients (22 male, 18 female) in 2 equal groups of 20 (11 male, 9 female) treated in a halo-vest with the two types of pins. Average age was 46 years (range 18–87 years). Diagnoses included traumatic cervical fractures (n = 34), pathologic cervical fracture (n = 1), cervical subluxations (n = 2), cervical chordoma (n = 1), Klippel-Feil syndrome (n = 1), and rheumatic disease (n = 1) (Table 1). During the 3-year, 2-month data collection period from 2000 to 2003, 179 halos were applied at the University of Michigan. Of these 179, 40 met our inclusion criteria: cervical trauma or pathology, use of airflow vest, Bremer crown application, use of 4 pin fixation (2 anterior, 2 posterior) with initial torque of 8 inch lbs (Figure 2). Additional inclusion criteria required the patient to be ambulatory, living in close enough proximity to the medical center that they could attend their follow-up appointments and be able to independently provide accurate verbal feedback while immobilized for a 12-week period. A number of patients were excluded from the study for the presence of spinal cord injury, attachment of thoracolumbosacral orthosis to halo, use of vest other than airflow, initial pin fixation of greater than 4 pins or less torque than 8 inch lbs, or if the fit of the vest or placement of the crown was deemed sub-optimal. Halos were placed by a certified member of the Orthotics staff, but inclusion in the study was determined by the principal investigator within 24 hours or by the time of the first follow-up visit.
Approval for the study was obtained through the university institutional review board, and patients signed an informed consent form at the time of initial application or by the first follow-up visit (4 weeks after initial halo application). Pin type selection was determined by the orthotics and prosthetics clinician at the time of initial application. Both skull pin types were inserted according to predetermined protocol: the halo crown is applied to the head with the use of positioning pins. The pin sites are determined, and considerations for pin torque and number of pins to be used are evaluated. Appropriate perpendicular orientation of the pins in relation to the skull is verified, and the pins are threaded through the crown. In 32 patients, a stab incision was placed before pin insertion, based on physician preference. Patients who received stab incisions were equally distributed in the two groups. The stab incision was used on the basis of physician preference, with rationale that it will minimize scarring and decrease skin impingement. All pins were then threaded further through the crown to break the skin or to penetrate the stab incision. The pins were tightened by hand in an opposing fashion: anterior right and posterior left at same time/anterior left and posterior right at same time. The pins were brought to “finger-tight” torque, followed by the use of an adjustable torque driver (set to 8 inch pounds) or a Bremer breakaway cap (preset to 8 inch pounds). The lock nuts were tightened hand-tight only, and the airflow vest was applied. The pins were retorqued after vest application to 8 inch pounds. All lock nuts were tightened with stabilization with the torque driver after initial retorque. No additional routine retorquing was done.
Patients were followed at intervals of 4 weeks throughout the 3-month period of immobilization. Instructions were given at the time of initial application and reviewed at each follow-up visit as necessary. Written instructions were given to increase the predictability of compliant behavior. Questionnaires were filled out at intervals of 24 to 48 hours, 3 to 4 weeks, 6 to 8 weeks, and 11 to 12 weeks after halo application. The clinical questionnaire was filled out by the orthotist and addressed our clinical evaluation of pin loosening. All evaluations were reviewed with the physician or orthopaedic staff at the time of visit. The pin sites were all examined for presence of swelling, erythema, and drainage or pin motion. If any indication of looseness was present, the lock nut was loosened and the pin was tested for lateral motion. Clinical loosening was defined as lateral motion of the pin with the lock nut loose as opposed to an isolated loss of torque. Retorquing did occur with the presentation of clinical loosening. Protocols for retorquing when looseness occurred were limited to two full revolutions of the pin. During the first 3 weeks after initial application, clinically loose pins were attempted to be retorqued to 6 inch pounds. After 3 weeks, clinically loose pins were attempted to be retorqued to 3 to 4 inch pounds only.
The definition of clinical loosening, originally set for the HIFix pins only, was applied to both types of pins for improved consistency in compiling results. If lateral motion was present, the pin was counted as loose. The need for retightening and presence of edema, erythema or drainage were also noted. If torque of 6 inch pounds was not achieved within the first 6 weeks and 3 to 4 inch pounds within the last 6 weeks, the pin was replaced. It was documented as a pin replacement and the pin site was evaluated for infection. The particular pin site involved was documented with the following stipulation: The attempt to determine which pin was originally problematic was not possible. The presence of a loose pin effectively creates an imbalance in the four-pin fixation system, resulting in multiple loosening. Therefore, it was documented which pin was loose and how many pins were loose; however, it was determined to not be an accurate data collection method or clinically relevant, with the exception of trends.
The subjective patient-directed questionnaire addressed the patient's personal experience including his or her perception of pin site pain, headache frequency, headache severity, sensation of movement at the pin sites, and sensation of pin pressure. A five-point scale was used to grade the answers (Table 2). A score of 1 was related to no pain, pressure, or motion; 5 was the maximum amount of pain, pressure, or motion. Answers were recorded directly from the patient. Issues of compliance were also addressed, including self-loosening or removing the vest, falling, bumping the halo, or self-adjusting the pins/locknuts. This did occur frequently to some degree in both groups, and a significant amount of time was spent reviewing instructions for proper pin and vest care. No gross compliance issues presented within either group, and, as a result, no patients were excluded for compliance issues from either group. Chi-square tests were performed to determine results for differences in pin loosening between the two types of pins.
Of the 40 patients, 20 patients were treated with standard halo pins and 20 patients were treated with HIFix pins. Six patients (30%) with HIFix pins had clinical loosening, whereas 13 patients (65%) with standard pins had clinical loosening within the 3-month immobilization period. Although this approached but did not reach statistical significance (p = 0.06), a definite trend was noted. One patient with HIFix pins (5%) went on to pin site infection and subsequent need for pin replacement. Four patients with standard pins (20%) went on to pin site infections, with a total of five pin replacements and one need for halo early removal.
All statistics for the subjective questionnaire showed higher quantitative scores (more discomfort) with the standard pins. Chi-square tests were performed to determine statistical significance in the five areas investigated: pin pressure (PPr), pin pain (PPa), headache frequency (HF), headache severity (HS), and sensation of pin movement (PM). There were no statistical differences present in four of the five areas at 24 hours after application: PPr (p = 0.66), PPA (p = 0.34), PM (p = 0.99), HS (p = 0.66). Headache severity did show a statistical significance at 24 hours after application (p = 0.02). At 4 weeks after application, differences between the HIFix pin and standard pins were present in statistical significance for all five areas PPa (p = 0.001), PPr (p = 0.001), HF (p = 0.001), HS (p = 0.001), and PM (p = 0.001). At 8-weeks post-application, HF (p = 0.001), PPr (p = 0.004), and PPa (p = 0.009) all presented with significant differences, favoring the HIFix pin. PM approached statistical significance (p = 0.055). By the end of the 12-week immobilization period, patients reported statistically significant differences between the two pins for PPa only (p = 0.018). Again, the HIFix pin was favorable. PPa and PPr were the most common complaints among patients. Significant differences were reported at 4 weeks (p = 0.001) and 8 weeks (p = 0.004) for PPr (Figure 3). PPa showed statistical significance at the intervals of 4 weeks (p = 0.001), 8 weeks (p = 0.009), and 12 weeks (p = 0.018) after initial halo application (Figure 4).
The results are based on the data collected from 40 patients over a 3-year, 2-month period. Length of time to collect this sample size is attributed to the attempt to eliminate multiple variables associated with halo placement and select a consistent population by which to compare. Voor and Khalily,14 in testing the biomechanical properties of the HIFix pin, showed that these changes in halo pin design affect the pin-bone interface by demonstrating less bony damage adjacent to the pin site and a decreased reliance on high axial forces to maintain fixation. This biomechanical study was performed on human cadaveric bone, with implications that the results may translate to fewer complications in a clinical setting. Other biomechanical studies have been performed to investigate methods of halo pin application.10,11,15,16 These studies indicate that variations in halo pin design, application force, and application angle may have a significant impact on the stability of the pin-bone interface. The majority of these studies were performed on cadaveric bone. Bullock and Runiciman,17 in 2001, looked at the biomechanical performance of these same two halo pin designs on bovine bone specimens with, and without, intact periosteum. Their results showed that the HIFix pin is biomechanically superior to the standard pin and that the HIFix pin was not affected negatively by the soft tissue as compared with the standard pin. Voor et al.,12 agreed with the conclusion that the HIFix pin is biomechanically superior and published results demonstrating that the 1-mm region of bone adjacent to the standard halo pin undergoes plastic deformation. This damaged adjacent bone leads to bony resorption, an enlarged pin site, and thus, loosening. The theory behind the design of the HIFix pin is that by decreasing this excess compressive load below the compressive yield strength of the skull, the damage to the bone adjacent to the pin will be reduced.
The probability value for the incidence of pin loosening was p = 0.06, approaching but not reaching statistical significance. However, although not statistically significant, a definite trend appears demonstrating that the HIFix pins performed better. Post hoc power analysis shows that increasing the sample size to 67 (from 40) would produce statistical significance for the incidence of pin loosening.
Activities of daily living such as standing up from a chair or simply leaning forward will assert pressure on the inferior edge of the vest. This superior-directed pressure places stress on the anterior pins and may be better resisted by the HIFix pin. The HIFix halo pins that did loosen were all after 8 weeks of immobilization. Standard pins loosened anywhere from 2 weeks to 12 weeks after initial application. This may be attributed to an ability of the HIFix pin to better withstand the forces applied throughout the halo during the first 8 weeks.
Regarding stab incisions, Botte et al.13 documented that the use of a stab incision at the time of initial pin insertion does not influence the stability and may lessen the infection rate. In their study, a halo skeletal fixator was studied prospectively in 51 patients, using two different pin insertion techniques, primarily the use of stab skin incisions versus straight pin insertion without stab skin incisions. There was no difference in infection, loosening, or resultant scars between the two groups.13 Our data agreed that there was no correlation between patients who had pin loosening and those who received a stab incision.
Our direct clinical comparison of the two types of halo pins demonstrated that the HIFix pin was superior and led to decreased pin loosening and subsequent complications. Isolating the multiple variances that exist with halo application as it relates to the incidence of pin loosening and other complications is difficult. This study demonstrates a significant trend, with the HIFix pin decreasing the incidence of loosening. Further testing with an increased sample size may be able to demonstrate a statistical difference, but the trend seems clear that there is decreased loosening with the HIFix pin.
The authors thank John McGillicuddy, MD, University of Michigan Department of Neurosurgery, for allowance of inclusion of neurosurgery patients, and David B. Weiss, MD, for editing assistance. The authors also acknowledge University of Michigan Acute Care Orthotics & Prosthetics staff for assistance in data collection and compliance to study protocols.
1. Perry J, Nickel V. Total cervical spine fusion for neck paralysis. J Bone Joint Surg
2. Johnson RM, Owen JR, Hart DL, Callahan RA. Cervical orthoses: a guide to their selection and use Clin Orthop
3. Koch RA, Nickel VL. The halo vest: an evaluation of motion and forces across the neck. Spine
4. Baum JA, Hanley EN, Pullekines J. Comparisons of halo complications in adults and children. Spine
5. Botte MJ, Byrne TP, Abrams RA, Garfin SR. Halo skeletal fixation techniques of application and prevention of complications. J Am Acad Orthop Surg
6. Victor DI, Bresnan MJ, Keller RB. Brain abscess complicating the use of halo traction. J Bone Joint Surg (Am)
7. Whitehill R, Richman JA, Glaser JA. Failure of immobilization of the cervical spine by the halo vest. a report of five cases. J Bone Joint Surg [Am]
8. Garfin SR, Botte MJ, Waters RL, Nickel VL. Complications in the use of the halo fixation device. J Bone Joint Surg [Am]
9. Mirza SK, Ross MR, Anderson PA, et al. Stabilizing properties of the halo apparatus. Spine
10. Botte MJ, Byrne TP, Garfin SR. Application of the halo device for immobilization of the cervical spine utilizing an increased torque pressure. J Bone Joint Surg [Am]
11. Garfin SR, Lee TQ, Roux RD, et al. Structural behavior of the halo orthosis pin-bone interface: biomechanical evaluation of standard and newly designed stainless steel halo fixation pins. Spine
12. Voor MJ, Anderson RC, Hart RT. Stress analysis of halo pin insertion by non-linear finite element modeling. J Biomech
13. Botte MJ, Byrne TP, Garfin SR. Use of skin incisions in the application of halo skeletal fixator pins. Clin Orthop Relat Res
14. Voor MJ, Khalily C. Halo pin loosening: a biomechanical comparison of experimental and conventional designs. J Biomech
15. Letts M, Girouard L, Yeadon A. Mechanical evaluation of four versus eight-pin halo fixation. J Pediatr Orthop
16. Triggs KJ, Ballock RT, Lee TQ, et al. The effect of angled insertion on halo pin fixation. Spine
17. Bullock SJ, Runciman J. Biomechanical evaluation of two halo pin designs, with, and without, intact periosteum. J Biomech