JBJS | Six-Pin Halo Fixation and the Resulting Prevalence of Pin-Site Complications

The Journal of Bone & Joint Surgery, Volume 83, Issue 3

Articles | March 01, 2001

Six-Pin Halo Fixation and the Resulting Prevalence of Pin-Site Complications

Jeffrey A. Nemeth, CO; L. George Mattingly, CO

Investigation performed at the Barrow Neurological Institute, Phoenix, Arizona
Jeffrey A. Nemeth, CO 1515 East Florence Boulevard, Suite 107, Casa Grande, AZ 85222. E-mail address: jeffnemeth@home.com
L. George Mattingly, CO 4445 North 7th Street, Phoenix, AZ 85014
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

The Journal of Bone & Joint Surgery. 2001; 83:377-377

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

Background: In spite of the many advances in halo application technique, the prevalence of complications associated with the use of halo fixation remains high, particularly at the pin sites. Many practitioners do not use more than four pins for halo application in adults because they believe that it increases the risk of complications. The purpose of this study was to investigate the use of six pins in halo application, in order to determine if the extra pins increased fixation strength without increasing the overall pin-site complication rate.

Methods: The first part of our study consisted of force-deflection tests conducted on models of the skull fitted with either a four or a six-pin halo to determine if the six-pin halo provided greater fixation strength. Each skull model was placed in a servocontrolled hydraulic test machine; an axial distraction force was then applied until failure occurred. The second part of the study was a retrospective analysis of sixty-three patient records to document the prevalence of pin-site complications in patients treated with a six-pin halo system; these findings were then compared with established complication rates associated with four-pin halos.

Results: In the force-deflection tests, the mean load to failure of the six-pin halo construct (2879 N [647 lb]) showed the system to be significantly stronger (p = 0.0033) than the four-pin halo construct (1681 N [378 lb]). Of the sixty-three patient records reviewed, five (8% [95% confidence interval, 1% to 15%]) revealed pin-loosening; no infection was recorded for these five patients. One of the sixty-three patients had redness and erythema at "multiple sites," but these areas healed well. Another presented with infection at all six sites; this was recorded as an allergic reaction.

Conclusions: Six-pin halo fixation results in greater halo strength and cervical spine stabilization without increasing the risk of pin-site complications.

Clinical Relevance: Our findings are relevant for current clinical practice as the high complication rates associated with halo application have deterred some practitioners from using this type of fixation. The use of six pins, along with an improved protocol for halo application and care, may contribute to a more successful treatment outcome with fewer complications.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

In the last forty years, stabilization of the cervical spine through halo fixation has evolved dramatically, with enhancement of the vest design and improvement of the protocol for application and follow-up^1-5. Yet the documented prevalence of complications due to various factors related to the halo remains high^1,6,7. The pin sites in particular seem to be associated with the greatest percentage of complications^1,6, with loose pins often leading to infection. Loose pins are the result of various factors, including a lack of sufficient fixation strength of the halo to withstand day-to-day jarring and falls. Recent studies have also shown that the compressive strength of the halo pin at each site diminishes greatly over time^8,9. The use of additional pins to secure the halo may contribute to greater fixation strength, as the overall compressive force of the halo is magnified, and therefore may reduce the risk of loosening (and infection).

The accepted theory, however, is that an increase in the number of pins increases the potential for complications⁷; it is this thinking that has deterred many practitioners from securing the halo with more than four pins. It is just as logical, however, to assume that an increase in the number of pins securing the halo would result in greater stabilization of the spine.

The following questions, therefore, remain unanswered: Does an increase in the number of halo pins result in an increase in fixation strength, and does the prevalence of complications (particularly at the pin sites) also increase? To answer these questions we first conducted a series of force-deflection trials on both four and six-pin halo constructs. Second, we retrospectively analyzed a series of patient records to determine the prevalence of complications with six-pin halos, focusing primarily on pin-site complications. These results were then compared with established complication rates for four-pin halos⁶. It was hypothesized that the use of additional pins would increase the fixation strength of the halo construct but would not increase the rate of pin-site complications.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1:: Biomechanical skull models with acetyl copolymer substrate located at the pin-substrate interfaces.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2:: Experimental setup of a biomechanical model in the hydraulic test machine.

View Larger

TABLE I: Ultimate Load to Failure

Four-Pin Trials			Six-Pin Trials
Trial	N	Lb	Trial	N	Lb
1	1560	351	1	2040	?459
2	1140	256	2	2750	?618
3	2480	558	3	3230	?726
4	2340	526	4	4470	1005
5	1390	312	5	2090	?470
6	1910	429	6	3370	?758
7	1020	229	7	2430	?546
8	1610	362	8	2650	?596
Mean	1681	378		2879	?647
Stand. dev.	?530	119		?800	?180

Add to My JBJS

TABLE I: Ultimate Load to Failure

Four-Pin Trials			Six-Pin Trials
Trial	N	Lb	Trial	N	Lb
1	1560	351	1	2040	?459
2	1140	256	2	2750	?618
3	2480	558	3	3230	?726
4	2340	526	4	4470	1005
5	1390	312	5	2090	?470
6	1910	429	6	3370	?758
7	1020	229	7	2430	?546
8	1610	362	8	2650	?596
Mean	1681	378		2879	?647
Stand. dev.	?530	119		?800	?180

View Larger

TABLE II: Comparison of Pin-Site Complication Rates Between Four and Six-Pin Halos*

*CI = confidence interval. †The complication rates associated with the four-pin halos were reported by Garfin et al.⁶.

Complication	Four-Pin Halos† (N = 179)			Six-Pin Halos (N = 63)
No. of Patients	Rate	95% CI	No. of Patients	Rate	95% CI
Pin-loosening	64	0.36	0.29-0.43	5	0.08	0.01-0.15
Pin-site infection	35	0.20	0.14-0.25	2	0.03	0-0.07
Pin-site bleeding	?2	0.01	0-0.03	0	0
Nerve injury	?3	0.02	0-0.04	0	0
Dural puncture	?1	0.01	0-0.02	0	0

Add to My JBJS

TABLE II: Comparison of Pin-Site Complication Rates Between Four and Six-Pin Halos*

*CI = confidence interval. †The complication rates associated with the four-pin halos were reported by Garfin et al.⁶.

Complication	Four-Pin Halos† (N = 179)			Six-Pin Halos (N = 63)
No. of Patients	Rate	95% CI	No. of Patients	Rate	95% CI
Pin-loosening	64	0.36	0.29-0.43	5	0.08	0.01-0.15
Pin-site infection	35	0.20	0.14-0.25	2	0.03	0-0.07
Pin-site bleeding	?2	0.01	0-0.03	0	0
Nerve injury	?3	0.02	0-0.04	0	0
Dural puncture	?1	0.01	0-0.02	0	0

View Larger

TABLE III: Comparison of Pin-Site Complication Rates in Historical Studies*

*CI = confidence interval.

Study	No. of Patients	Mean Age (range) (yr)	Loosening			Infection			Penetration			Loss of Position or Dislodgment			Change in Protocol
No.	Rate	95% CI	No.	Rate	95% CI	No.	Rate	95% CI	No.	Rate	95% CI
Nickel et al.⁷, 1968	204	17 (4-52)	204	1.00	1.00	?5	0.20	0.15-0.25
Garfin et al.⁶, 1986	179	28.3 (2-90)	?64	0.36	0.29-0.43	35	0.20	0.14-0.25	1	0.01	0-0.02
Botte et al.⁴, 1987	?42	36 (16-70)	??3	0.07	0-0.15	?1	0.02	0-0.06							Increased torque from 0.68 Nm to 0.90 Nm
Lind et al.¹⁶, 1988	?83	38 (13-89)	?50	0.60	0.49-0.71	18	0.22	0.13-0.31	1	0.01	0-0.03	11	0.13	0.06-0.20
Baum et al.¹⁹, 1989	?93	(5-83)	?15	0.16	0.09-0.23	?8	0.09	0.03-0.15	1	0.01	0-0.03	1	0.01	0-0.03	Pin placement based on study by Garfin et al.⁶
Vertullo et al.¹⁷, 1997	266	35.3 (11-93)	?10	0.04	0.02-0.06	16	0.06	0.03-0.09	1	0.01	0-0.02	7	0.03	0.01-0.05	Routine pin-retightening at 24 hr and 1 wk
Current study	?63	40.5 (16-85)	??5	0.08	0.01-0.15	?2	0.03	0-0.07	0	0		0	0		6-pin application

Add to My JBJS

TABLE III: Comparison of Pin-Site Complication Rates in Historical Studies*

*CI = confidence interval.

Study	No. of Patients	Mean Age (range) (yr)	Loosening			Infection			Penetration			Loss of Position or Dislodgment			Change in Protocol
No.	Rate	95% CI	No.	Rate	95% CI	No.	Rate	95% CI	No.	Rate	95% CI
Nickel et al.⁷, 1968	204	17 (4-52)	204	1.00	1.00	?5	0.20	0.15-0.25
Garfin et al.⁶, 1986	179	28.3 (2-90)	?64	0.36	0.29-0.43	35	0.20	0.14-0.25	1	0.01	0-0.02
Botte et al.⁴, 1987	?42	36 (16-70)	??3	0.07	0-0.15	?1	0.02	0-0.06							Increased torque from 0.68 Nm to 0.90 Nm
Lind et al.¹⁶, 1988	?83	38 (13-89)	?50	0.60	0.49-0.71	18	0.22	0.13-0.31	1	0.01	0-0.03	11	0.13	0.06-0.20
Baum et al.¹⁹, 1989	?93	(5-83)	?15	0.16	0.09-0.23	?8	0.09	0.03-0.15	1	0.01	0-0.03	1	0.01	0-0.03	Pin placement based on study by Garfin et al.⁶
Vertullo et al.¹⁷, 1997	266	35.3 (11-93)	?10	0.04	0.02-0.06	16	0.06	0.03-0.09	1	0.01	0-0.02	7	0.03	0.01-0.05	Routine pin-retightening at 24 hr and 1 wk
Current study	?63	40.5 (16-85)	??5	0.08	0.01-0.15	?2	0.03	0-0.07	0	0		0	0		6-pin application

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

Force Distraction

Our first objective was to determine the fixation strength (load to failure) of the six-pin halo construct through a series of force-deflection tests conducted on biomechanical models of the human skull.

Skull Models

To create our biomechanical skull models, we first made an alginate impression of a human skull with a normal anatomy. This allowed us to create a mold duplicating the human skull, which was then filled with plaster of Paris; sixteen models were created from the same mold. The plaster models were then blister-formed with 3/8-in (0.95-cm) thick Durplex (Piper Plastics, Chandler, Arizona), which was extended around the base of the model (Fig. 1). This extra edge allowed for greater stabilization of each model in the distraction unit and facilitated the placement of clamps for securing the model (Fig. 2). The combination of the plaster and Durplex also created a model with a weight similar to the mean weight of the human head.

To simulate the pin-bone interface achieved in halo application, a substrate of 25% glass fiber-filled acetyl copolymer (Piper Plastics) was incorporated into the model at six circular sites (Fig. 1). This allowed us to create a pin site with interface characteristics similar to the reaction of pin and bone. This experimental design and this model were used successfully in an earlier study of halo-fixation strength¹⁰.

Halo Application and Pin Placement

Each skull model was fitted with an Ace titanium closed-back halo ring, attached with either four or six 2-in (5-cm) Ace titanium pins (Ace Medical Company, Los Angeles, California) at the model-substrate interface. We selected the Ace halo ring because we thought that, of all of the current rings available, it most closely approximates the shape of the skull, thereby allowing a more accurate pin-bone interface. Each halo was applied to the skull model according to an established protocol: perpendicular pin insertion tightened to a measurement of 0.90 Nm (8 in-lb) of torque^1,2. The pins were tightened with use of a load-cell to digitally calibrate the level of torque in order to ensure and gauge the accuracy of application, as industry-standard torque wrenches have been found to be inherently inaccurate¹¹.

In the four and six-pin halo constructs, the two anterior pins were placed in similar positions within the safe zone for pin placement. This safe zone is defined as "the anterolateral aspect of the skull, approximately 1 cm superior to the orbital rim, cephalad to the lateral two thirds of the orbit, and below the greatest circumference of the skull."^1,12-14 In this position, the pins do not interfere with the temporalis muscle or with the zygomaticotemporal nerve and are not in danger of penetrating the sinuses. There are not the same neuromuscular restrictions with regard to the placement of the posterior pins, and skull thickness does not vary as greatly posteriorly^13,14. Therefore, we selected a site 1 cm posterior to each model "ear." In the six-pin halo constructs, an additional two pins were placed 2 cm posterior to the "ear."

Load to Halo Failure by Distraction

Eight trials were conducted on the four-pin halo constructs, and eight were performed on the six-pin halo constructs. Each skull model was mounted in the test frame (Fig. 2), and an axial distraction force simulating traction of the halo was applied at a constant ramp rate of 1 mm/sec with use of a servocontrolled hydraulic test machine (Shore Western Material Test System; Shore Western Manufacturing, Monrovia, California) until failure occurred at the pin-substrate interface. The load to failure was measured with a load-cell mounted on the hydraulic test machine. Failure was defined as a decrease in load, confirmed by visible movement at the pin-substrate interface.

Review of Patient Records

Patient records concerning halo applications performed from December 1994 through December 1997 at the Barrow Neurological Institute in Phoenix, Arizona, were examined. All adult patients treated with a halo during this time-period were managed with a six-pin halo system unless the specific injury prohibited the use of six pins. The requirements for inclusion in our retrospective study were as follows: (1) the patient had to have worn the halo for a minimum of two weeks, (2) the hospital charts needed to be complete (including documentation of follow-up) and available, and (3) the patient had to be a minimum of sixteen years old, in order to ensure appropriate skull development¹⁰. The same technique was used for the application of all of the halos, and all of the halo vests were made by the same manufacturer (IQM, Tempe, Arizona) as were all of the superstructures, rings, and pins (Ace Medical Company). The titanium closed-back halo ring and 2-in (5-cm) titanium pins used in the treatment of these patients were the same as those used in our force-deflection study.

The records of sixty-three patients were considered to have sufficient documentation for inclusion in the study. Forty-three (68%) of the patients were male, and twenty (32%) were female. The mean age at the time of halo application was 40.5 years, the median age was thirty-nine years, the mode was eighteen years, and the age range was sixteen to eighty-five years. Forty-nine (78%) of the patients were neurologically intact at the time of the halo application.

The majority of the patients (fifty; 79%) required halo treatment because of trauma to the cervical spine (caused by a motor-vehicle accident or a gunshot wound). Four other patients required the halo because of instability due to complications from arthritis. Of the remaining nine halos, two were used for patients with Arnold-Chiari malformation; three, central cord syndrome; two, a lytic lesion; and one, postoperative angulation. One patient record did not list a clear diagnosis requiring the use of the halo.

Comparison of Complication Rates with Those Established for Four-Pin Configuration

Our objective in analyzing the sixty-three patient records was to establish complication rates for six-pin halo application, which could then be compared with published complication rates for four-pin halo application. For the purposes of our comparison, we employed the method used in the 1986 study by Garfin et al.⁶ for review of complication rates and for definition of categories of pin-site complications. Although the data in that study are somewhat dated, they are the most complete data available and constitute the benchmark that almost all subsequent investigators have used to compare complication rates when studying independent effects of changes to halo protocol such as pin-retightening or angle of pin insertion.

Our definition of pin-loosening, however, differed from that used by Garfin et al., who documented loosening "when a pin could be freely twisted by the examiner without resistance, or the tip of the pin was visible at the edge of the skin, rather than being secured against the skull."6 We thought that any indication of looseness could potentially lead to a complication because, as Garfin et al. pointed out, pin-loosening is often a forerunner to infection. In our study and in clinical practice, we expanded the definition of pin-loosening to include any measurement of <0.90 Nm (<8 in-lb) of torque at the pin site; therefore, any mention of pin-loosening in the patient records was included.

Results

Introduction | Materials and Methods | Results | Discussion | References

Force Distraction

The mean load to failure (and standard deviation) of the four-pin halo constructs was 1681 ± 530 N (378 ± 119 lb), and the range was 1020 to 2480 N (Table I). The mean load to failure of the six-pin halo constructs was 2879 ± 800 N (647 ± 180 lb), and the range was 2040 to 4470 N. The load to failure of the six-pin constructs was 171% of that of the four-pin constructs. The two-tailed t test demonstrated that the six-pin halo model provided significantly greater fixation strength than the four-pin model (p = 0.0033).

All failures occurred at the pin-substrate interface. No model was visibly distorted or damaged; thus, we were confident that the design of the model was a controlled variable that did not influence our results.

A post hoc power study of our results indicated that the power to detect an increase in fixation strength of 171% was approximately 84%; the minimum required sample size for this level of power was four. We initially conducted four trials of each type of halo construct; however, our findings did not produce significant results or allow for us to either accept or reject the null hypothesis. We therefore conducted four additional trials, which produced a highly significant difference in fixation strength with the use of the additional pins.

Review of Patient Records

Pin-Loosening

Pin-loosening was recorded for five (8%) of the sixty-three patients; infection was not recorded for any of the five patients, although infection was noted in two other patients (Table II). Only four of the five patient records specified how many pins were loose: a total of seven pins were loose in these four patients. Two pins in the anterior position, one in the mid-posterior position, and four in the posterior position were loose. The 2.4% prevalence of pin-loosening in our study (nine of 378 pins) was considerably lower than the prevalence in the study by Garfin et al.⁶ (25% [180] of 716 pins). (On the basis of the best-estimate measurement [the mean number of loose pins per patient], the patient for whom the record did not specify the number of loose pins in our study was considered to have two loose pins.)

Pin-Site Infection

One patient was noted to have redness and erythema at multiple sites with one drop of discharge, but the sites were noted to have healed well. One patient presented with all six sites infected, and the patient’s records indicated that the infection was due to an allergic reaction to the titanium in the pins. Neither patient had a deep infection or one that resulted in osteomyelitis or a subdural abscess. The percentage of pin sites with infection in our study (1.9% [seven] of 378) was also lower than the percentage of infected sites reported by Garfin et al.⁶ (9.4% [sixty-seven] of 716). (In the patient with redness and erythema in our study, only the pin site with the discharge was considered to be infected.)

Other Complications

We did not find evidence of any other pin-site complications in our study. This observation is consistent with the finding of Garfin et al.⁶ that complications such as bleeding, nerve injury, and dural puncture are rare.

Discussion

Introduction | Materials and Methods | Results | Discussion | References

In our study of the complications that can occur when a six-pin halo construct is used, we chose to concentrate on those occurring at the pin-bone interface. Our reason was twofold. First, when a halo fails (must be removed) for reasons related to the integrity of the apparatus, as opposed to patient factors (such as an injury that makes it impossible to maintain acceptable cervical reduction or to achieve spinal stability¹⁴), the failure usually occurs at the pin-bone interface. According to Botte et al., "failure of the halo at the rod-ring or rod-vest interface is rare."¹ When the pins become loose, the fixation strength of the halo (and its ability to stabilize the spine) is compromised. This loosening almost always precedes infection.

Second, an increased risk of pin-site complications has been the conventional argument against the use of additional pins for halo application in adults. The number and type of pins used in halo application have remained relatively unchanged since the introduction of the halo device by Perry and Nickel in 1959⁵. In 1968, Nickel et al. again validated the use of four pins by stating that: "When we attempted to increase the rigidity of fixation by using eight pins instead of four, it aggravated skin damage and precipitated premature removal of the halo."⁷ This statement effectively closed the door on increasing the number of pins used in clinical practice.

The data from our retrospective clinical review showed no increase in the number of pin-site complications with the use of six pins: in fact, the prevalence in our study was less than that reported by Garfin et al. for a four-pin construct⁶. In addition, our force-deflection trials on adult skull models demonstrated significantly higher strength with six-pin halo fixation, almost double (171%) the strength of the four-pin halo.

Most other studies on increased pin usage have concentrated primarily on pediatric application, as a child’s skull can vary greatly in thickness and experiences more jarring and microtrauma during everyday activities¹⁵. Micromotion forces at each pin site, therefore, are increased and must be addressed. The focus has been on using more pins with less torque at each site to reduce the risk of penetration and to achieve the same stability. One recent study compared four-pin with eight-pin halo fixation of the same torque and showed that the eight-pin system increased the stiffness of the pins by 24.2%¹⁵. The authors of that study suggested that, while the overall stiffness of the halo was not greatly improved by the use of additional pins, the maximum load to pin slippage might be "quite higher." Our force-deflection study complements this research by proving that the load to pin slippage (or failure) is indeed significantly higher when additional pins are used. Fleming et al.^8,9, in similar studies on four patients, demonstrated that the compressive forces at the pin-bone interface of the posterior halo pins decreased by more than 80% over a period of three months. The extra strength provided by employing six pins, therefore, could expand the clinical usefulness of the halo.

Our decision to use six pins (rather than more) was based on clinical experience; our purpose was to balance improved stability with neuromuscular, cosmetic, and patient-comfort considerations. Additional pins are not easily placed in the anterior positions without compromising the sinuses or existing musculature¹. The two anterior halo pins have also been shown to be slightly more susceptible to loosening and pin-site complications¹³. According to Garfin et al.⁶, "fifty-three per cent of the pins that became loose were located anteriorly" and "sixty per cent of the infections were observed around the anterior pins." One hypothesis is that interface forces are greater in these anterior positions. The logical placement of additional pins, therefore, is posterior, where the skull is thicker and there are fewer neuromuscular considerations¹.

Although the rate of pin-site complications in our study was lower than that reported by Garfin et al.⁶, it cannot be assumed that the use of the additional pins caused the lower prevalence. A historical review of studies on complication rates associated with halo use (Table III) showed a general trend toward decreasing rates with improvements in application protocol. In the original study by Nickel et al., 100% of the patients had loose pins⁷. The rate of 8% in our study is more in keeping with the findings of other current studies in which contemporary protocols were used. The halos in our study were all applied with use of 0.90 Nm (8 in-lb) of torque. A torque of 0.90 Nm results in mechanical qualities that are greatly superior to those provided by a torque of 0.68 Nm¹. In addition, it has been shown that increasing the amount of torque from 0.68 Nm to 0.90 Nm decreases the prevalence of pin-loosening from 36% to 7% and reduces the rate of infection from 20% to 2%^1,4. In choosing the pin sites for application of the halos, we also followed documented recommendations for pin placement: "The anterolateral aspect of the skull, approximately 1 cm superior to the orbital rim, cephalad to the lateral two thirds of the orbit, and below the greatest circumference of the skull."¹ Improper pin placement, especially in a position that is too high on the head, can lead to loss of position or dislodgment of the halo^16,17.

Furthermore, we believe that the distance between the halo ring and the skull is a factor affecting complications. Our goal is to have as close a fit between the ring and the skull as possible, while still allowing for proper cleaning of the pin sites. A close-fitting ring decreases the lever arm of the pin, which in turn reduces the moment of torque acting on the pin site and the subsequent motion at the pin-bone interface.

As the components in the halo vest do not act independently of each other, the technique for vest application and the type of vest that is used may also have contributed to our lower complication rates. An important factor in pin-loosening may be the magnitude of the loads applied to the pins by forces transmitted from the vest to the halo³. An improperly fit or poorly designed vest may lead to pin-loosening and therefore to infection¹⁸.

Finally, better follow-up care may have played an important role in our study. A 1997 study by Vertullo et al. showed that an active, prescribed program of follow-up care, including pin-retightening at twenty-four hours and again at one week after application, resulted in fewer complications¹⁷. Other studies have also shown that active follow-up care contributes to lower complication rates^14,19,20.

Our data are limited in several ways. The force-deflection tests were a laboratory-controlled demonstration of only one type of force acting on the pins; they did not account for shear forces. They also did not account for dynamic forces associated with tendon or muscle activity, which can also create motion and loosening at the pin site. In our test results, the variation in the load to failure among the halos was likely due to the placement of the halo pins in slightly different locations within a 2.5-cm circle (the pin-substrate interface). The skull model also cannot account for dynamic physiological processes that may contribute to clinical failure of the halo, including bone resorption, necrosis, and remodeling. Our retrospective analysis of patient records was also subject to variations in the amount of information provided in each record as well as to practitioner-introduced variation in halo application (although the procedure for application was the same for all patients).

Our force-deflection study demonstrated that halo fixation strength may be increased through the use of six pins, while the findings of our retrospective study disproved the belief that the additional pins increase the prevalence of complications and halo failure. It may be inferred, in fact, that the use of six pins as part of a prescribed protocol for halo application may actually reduce the likelihood of pin-site complications (particularly loosening and, therefore, infection) as there may be reduced motion at the pin-sites. These findings are relevant for current clinical practice as the high complication rates with halo applications have deterred some practitioners from using them.

References

Introduction | Materials and Methods | Results | Discussion | References

Botte MJ; Byrne TP; Abrams RA; and Garfin SR: Halo skeletal fixation: techniques of application and prevention of complications. J Am Acad Orthop Surg,1996.4: 44-53, 444 1996 [PubMed]

Triggs KJ; Ballock RT; Lee TQ; Woo SL; and Garfin SR: The effect of angled insertion on halo pin fixation. Spine,1989.14: 781-3, 14781 1989 [PubMed]

Krag MH, and Beynnon BD: A new halo-vest: rationale, design, and biomechanical comparison to standard halo-vest designs. Spine,1988.13: 228-35, 13228 1988 [PubMed]

Botte MJ; Byrne TP; and Garfin SR.: Application of the halo device for immobilization of the cervical spine utilizing an increased torque pressure. J Bone Joint Surg Am,1987.69: 750-2, 69750 1987 [PubMed]

Perry J, and Nickel VL: Total cervical-spine fusion for neck paralysis. J Bone Joint Surg Am,1959.41: 37-59, 4137 1959 [PubMed]

Garfin SR; Botte MJ; Waters RL; and Nickel VL: Complications in the use of the halo fixation device. J Bone Joint Surg Am,1986.68: 320-5, 68320 1986 [PubMed]

Nickel VL; Perry J; Garrett A; and Heppenstall M: The halo. A spinal skeletal traction fixation device. J Bone Joint Surg Am,1968.50: 1400-9, 501400 1968 [PubMed]

Fleming BC; Huston DR; Krag MH; and Sugihara S: Pin force measurement in a halo-vest orthosis, in vivo. J Biomech,1998.31: 647-51, 31647 1998 [PubMed]

Fleming BC; Krag MH; Huston DR; and Sugihara S.: Pin loosening in a halo-vest orthosis: a biomechanical study. Spine,2000.25: 1325-31, 251325 2000 [PubMed]

Lerman JA; Haynes RJ; Koeneman EJ; Koeneman JB; and Wong WB: A biomechanical comparison of Gardner-Wells tongs and halo device used for cervical spine traction. Spine,1994.19: 2403-6, 192403 1994 [PubMed]

Smith MD; Johnson LJ; Perra JH; and Rawlins BA: A biomechanical study of torque and accuracy of halo pin insertional devices. J Bone Joint Surg Am,1996.78: 231-8, 78231 1996 [PubMed]

Wong WB, and Haynes RJ: Osteology of the pediatric skull. Considerations of halo pin placement. Spine,1994.19: 1451-4, 191451 1994 [PubMed]

Garfin SR; Botte MJ; Centeno RS; and Nickel VL: Osteology of the skull as it affects halo pin placement. Spine,1985.10: 696-8, 10696 1985 [PubMed]

Marciano FF; Greene KA; Mattingly LG; and Dickman CA: Halo brace immobilization of the cervical spine: A review of principles and application techniques. BNI Quarterly,1994.10: 13-7, 1013 1994

Letts M; Girouard L; and Yeadon A: Mechanical evaluation of four- versus eight-pin halo fixation. J Pediatr Orthop,1997.17: 121-4, 17121 1997 [PubMed]

Lind B; Sihlbom H; and Nordwall A: Halo-vest treatment of unstable traumatic cervical spine injuries. Spine,1988.13: 425-32, 13425 1988 [PubMed]

Vertullo CJ; Duke PF; and Askin GN: Pin-site complications of the halo thoracic brace with routine pin re-tightening. Spine,1997.22: 2514-6, 222514 1997 [PubMed]

Walker PS; Lamser D; Hussey RW; Rossier AB; Farberov A; and Dietz J: Forces in the halo-vest apparatus. Spine,1984.9: 773-7, 9773 1984 [PubMed]

Baum JA; Hanley EN Jr; and Pullekines J: Comparison of halo complications in adults and children. Spine,1989.14: 251-2, 14251 1989 [PubMed]

Botte MJ; Byrne TP; Abrams RA; and Garfin SR: The halo skeletal fixator: current concepts of application and maintenance. Orthopedics,1995.18: 463-71, 18463 1995 [PubMed]

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1:: Biomechanical skull models with acetyl copolymer substrate located at the pin-substrate interfaces.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2:: Experimental setup of a biomechanical model in the hydraulic test machine.

View Larger

TABLE I: Ultimate Load to Failure

Four-Pin Trials			Six-Pin Trials
Trial	N	Lb	Trial	N	Lb
1	1560	351	1	2040	?459
2	1140	256	2	2750	?618
3	2480	558	3	3230	?726
4	2340	526	4	4470	1005
5	1390	312	5	2090	?470
6	1910	429	6	3370	?758
7	1020	229	7	2430	?546
8	1610	362	8	2650	?596
Mean	1681	378		2879	?647
Stand. dev.	?530	119		?800	?180

Add to My JBJS

TABLE I: Ultimate Load to Failure

Four-Pin Trials			Six-Pin Trials
Trial	N	Lb	Trial	N	Lb
1	1560	351	1	2040	?459
2	1140	256	2	2750	?618
3	2480	558	3	3230	?726
4	2340	526	4	4470	1005
5	1390	312	5	2090	?470
6	1910	429	6	3370	?758
7	1020	229	7	2430	?546
8	1610	362	8	2650	?596
Mean	1681	378		2879	?647
Stand. dev.	?530	119		?800	?180

View Larger

TABLE II: Comparison of Pin-Site Complication Rates Between Four and Six-Pin Halos*

*CI = confidence interval. †The complication rates associated with the four-pin halos were reported by Garfin et al.⁶.

Complication	Four-Pin Halos† (N = 179)			Six-Pin Halos (N = 63)
No. of Patients	Rate	95% CI	No. of Patients	Rate	95% CI
Pin-loosening	64	0.36	0.29-0.43	5	0.08	0.01-0.15
Pin-site infection	35	0.20	0.14-0.25	2	0.03	0-0.07
Pin-site bleeding	?2	0.01	0-0.03	0	0
Nerve injury	?3	0.02	0-0.04	0	0
Dural puncture	?1	0.01	0-0.02	0	0

Add to My JBJS

TABLE II: Comparison of Pin-Site Complication Rates Between Four and Six-Pin Halos*

*CI = confidence interval. †The complication rates associated with the four-pin halos were reported by Garfin et al.⁶.

Complication	Four-Pin Halos† (N = 179)			Six-Pin Halos (N = 63)
No. of Patients	Rate	95% CI	No. of Patients	Rate	95% CI
Pin-loosening	64	0.36	0.29-0.43	5	0.08	0.01-0.15
Pin-site infection	35	0.20	0.14-0.25	2	0.03	0-0.07
Pin-site bleeding	?2	0.01	0-0.03	0	0
Nerve injury	?3	0.02	0-0.04	0	0
Dural puncture	?1	0.01	0-0.02	0	0

View Larger

TABLE III: Comparison of Pin-Site Complication Rates in Historical Studies*

*CI = confidence interval.

Study	No. of Patients	Mean Age (range) (yr)	Loosening			Infection			Penetration			Loss of Position or Dislodgment			Change in Protocol
No.	Rate	95% CI	No.	Rate	95% CI	No.	Rate	95% CI	No.	Rate	95% CI
Nickel et al.⁷, 1968	204	17 (4-52)	204	1.00	1.00	?5	0.20	0.15-0.25
Garfin et al.⁶, 1986	179	28.3 (2-90)	?64	0.36	0.29-0.43	35	0.20	0.14-0.25	1	0.01	0-0.02
Botte et al.⁴, 1987	?42	36 (16-70)	??3	0.07	0-0.15	?1	0.02	0-0.06							Increased torque from 0.68 Nm to 0.90 Nm
Lind et al.¹⁶, 1988	?83	38 (13-89)	?50	0.60	0.49-0.71	18	0.22	0.13-0.31	1	0.01	0-0.03	11	0.13	0.06-0.20
Baum et al.¹⁹, 1989	?93	(5-83)	?15	0.16	0.09-0.23	?8	0.09	0.03-0.15	1	0.01	0-0.03	1	0.01	0-0.03	Pin placement based on study by Garfin et al.⁶
Vertullo et al.¹⁷, 1997	266	35.3 (11-93)	?10	0.04	0.02-0.06	16	0.06	0.03-0.09	1	0.01	0-0.02	7	0.03	0.01-0.05	Routine pin-retightening at 24 hr and 1 wk
Current study	?63	40.5 (16-85)	??5	0.08	0.01-0.15	?2	0.03	0-0.07	0	0		0	0		6-pin application

Add to My JBJS

TABLE III: Comparison of Pin-Site Complication Rates in Historical Studies*

*CI = confidence interval.

Study	No. of Patients	Mean Age (range) (yr)	Loosening			Infection			Penetration			Loss of Position or Dislodgment			Change in Protocol
No.	Rate	95% CI	No.	Rate	95% CI	No.	Rate	95% CI	No.	Rate	95% CI
Nickel et al.⁷, 1968	204	17 (4-52)	204	1.00	1.00	?5	0.20	0.15-0.25
Garfin et al.⁶, 1986	179	28.3 (2-90)	?64	0.36	0.29-0.43	35	0.20	0.14-0.25	1	0.01	0-0.02
Botte et al.⁴, 1987	?42	36 (16-70)	??3	0.07	0-0.15	?1	0.02	0-0.06							Increased torque from 0.68 Nm to 0.90 Nm
Lind et al.¹⁶, 1988	?83	38 (13-89)	?50	0.60	0.49-0.71	18	0.22	0.13-0.31	1	0.01	0-0.03	11	0.13	0.06-0.20
Baum et al.¹⁹, 1989	?93	(5-83)	?15	0.16	0.09-0.23	?8	0.09	0.03-0.15	1	0.01	0-0.03	1	0.01	0-0.03	Pin placement based on study by Garfin et al.⁶
Vertullo et al.¹⁷, 1997	266	35.3 (11-93)	?10	0.04	0.02-0.06	16	0.06	0.03-0.09	1	0.01	0-0.02	7	0.03	0.01-0.05	Routine pin-retightening at 24 hr and 1 wk
Current study	?63	40.5 (16-85)	??5	0.08	0.01-0.15	?2	0.03	0-0.07	0	0		0	0		6-pin application