Three Methods Of Guided Growth For Pediatric Lower Extremity Angular Deformity Correction

Abstract

Background

Different methods of guided growth are used for correction of angular deformity in growing children. The differences between these different methods are not well described in the literature.

Methods

A retrospective review was undertaken comparing the effectiveness and complication rates of titanium staples, titanium eight-plates, and the stainless steel Pedi-plate at a tertiary pediatric hospital after IRB approval.

Results

77 patients were included in the analysis. Average follow up was 18 months after implantation (range 7-22). Stainless steel implants showed significantly lower complication rate compared to the other groups with significantly faster rate of deformity correction when compared to titanium staples.

Conclusion

Our data can be used to guide implant choices for guided growth.

Introduction

Angular deformity of the lower extremity in the coronal plane may be the result of many etiologies including exaggerated physiologic angulation, trauma, infection, skeletal dysplasia, and metabolic disease¹. Deformities can be unacceptable in appearance, affect gait, cause knee pain, alter knee biomechanics leading to ligamentous instability, and may be detrimental to long term joint function. No long term study has correlated angular deformity with joint degeneration but it has been implicated in placing larger than normal loads across the joints predisposing to cartilage breakdown and joint degeneration². Treatment of a growing child’s angular deformity to avoid these sequelae via manipulation of the physis is a well-accepted treatment method¹,³-⁵.

chniques utilized to manipulate physeal growth in order to achieve angular correction via guided growth have changed over time. The permanent partial physeal arrest was initially described over 70 years ago but requires specific timing and can lead to over or under correction due to the unpredictable nature of skeletal growth. Temporary, reversible techniques using staples, plates, or percutaneous screws have been developed to place more control over the correction into the hands of the surgeon¹,⁶.These devices are not without complications, related to both the technique as well as the hardware. Staples have been used successfully for many years but there have been many reports of breakage, backing out, malpositioning, and difficulties with placement and removal leading to possible irreversible physeal arrest¹,⁵,⁷. More recently plate and screw devices have been developed to allow for more reliable and secure implant positioning, utilizing only one implant. Furthermore, a biomechanical advantage has been proposed to focus the forces more peripherally than a fixed angle staple improving the working distance⁴,⁷. Hardware breakage has also been reported with these devices, leading to several design changes over time⁴,⁷,⁸.

Surgeons at our institution have utilized several different temporary guided growth techniques. The purpose of this study is to document angular deformity correction and complications in patients undergoing guided growth treatment at our institution. We hypothesized that these devices provided similar amounts of radiographic angular correction with fewer complications and revision surgeries with the newest device, a stainless steel guided growth plate.

Materials and Methods

This data was collected in a retrospective manner, identifying patients by a query of operating room implant utilization databases following Institutional Review Board approval. We analyzed the use of three different implants, the titanium hemiepiphyseal staple (Smith and Nephew, Inc., Memphis, TN) (Group S), the titanium eight-plate (Orthofix Inc., Lewisville, TX) (Group E), and the stainless steel Pedi-plate (Orthopedatrics Inc., Warsaw, IN) (Group P). A total of 77 patients were identified from 1999 through 2010 who met our inclusion criteria which included any patient with a coronal plane angular deformity about the knee, secondary to any underlying diagnosis, with at least six months of clinical follow up. We excluded patients who had concomitant procedures during the initial six months that would affect the angular and mechanical axis measurements or did not have preand post-implantation full length standing radiographs

The mechanical tibiofemoral angle (MTFA), lateral distal femur angle (LDFA), and medial proximal tibial angle (MPTA) were measured as described by Paley⁹. Early in the study, the mechanical axis and anatomic angles were measured by hand with goniometer by a senior orthopedic resident. After implementation of a digital imaging system, these angles were determined by utilizing digital software (OrthoView LLC, Jacksonville, FL). Statistics were calculated using SPSS software with a significance level specified at p<0.05. ANOVA with subsequent post-hoc Tukey HSD analysis was performed for any significant relationships as well as Chi-square analysis.

Results

A total of 77 patients were identified that met our inclusion criteria, these patients underwent a total of 188 instrumentations (97 distal femur physis procedures, 91 proximal tibial physis procedures). Average follow up for all patients from the initial surgery was 18 months

Descriptive Analysis

Descriptive information for the population and each group is included in Tables I and Tables II.

Table 1.

Descriptive Results

	All	Group S	Group E	Group P
Number of Patients	77	18	24	43
Number of Physes Treated	188 (avg 2.44)	47 (avg 2.61)	55 (avg 2.29)	86 (avg 2.0)
Average Follow-up Months	18	7	22	13
Average Age at procedure	11.64	10.91 (5.9-15.1)	11.72 (6.2-19)	11.46 (2.4-16.7)
Average BMI at procedure	30.24	31.49	30.29	29.71
Top diagnoses by percentage
Blount’s %	31.6	31.9	40	14
Idiopathic genu valgum %	23.2	2.1	12.7	32.6
Hypophoshatemic Rickets %	20	8.5	25.5	11.6
Epiphyseal Dysplasia %	9.7	23.4	7.3	11.6

Group S: Staples, Group E: Titanium eight-plate, Group P: Stainless steel Pedi-plate

Table II.

Angular correction

	All	Group S	Group E	Group P
MTFA:
Pre op	15.8	18.5	14.1	15.4
Post op	10.76	12	11	6.4
Change	5.1	6.5	3.1	6.4
Correction per year in degrees	3.73	4.16	2.09	4.34
*: statistically significant with p:0.028 (between groups E and P)

		All	Group S	Group E	Group P
Valgus	LDFA: Pre op Post top change	81 87 5.6	80 87 7	81 84 3.1	87 81 5.5
	MPTA: Pre op Post top change	94 91 2.94	94 92 2.1	92 93 1.29	94 90 4.33
Varus	LDFA: Pre op Post top change	95 91 3.7	98 94 4	94 90 3.7	95 91 4.2
	MPTA: Pre op Post top change	77 81 3.25	75 79 4	79 82 3.17	79 82 4
#: statistically significant with p:0.010 (between groups E and P)

MTFA: Mechanical Tibiofemoral Angle, LDFA: Lateral Distal Femoral Angle, MPTA: Medial Proximal Tibial Angle, Group S: Staple, Group E: Titanium eight-plate, Group P: Stainless steel Pedi-plate

Angular Correction Analysis

Analysis of the angular correction based on the radiographic measurements, including the MTFA, LDFA, and MPTA are summarized in Tables III. Absolute values of MTFA were used for analysis in order to combine the data from both varus and valgus deformities. Analysis of the angular correction obtained did show that the MTFA change between the groups was significant. Post hoc analysis revealed that the correction in Group S was significantly more than that obtained in Group E (3.81 degrees, p = 0.028, 3.81). The correction obtained in Group P compared to Group E approached statistical significance (2.97 degrees, p = 0.056). A separate analysis was performed grouping the patients into varus and valgus preoperative deformity and comparing LDFA and MPTA (Tables III). This revealed that in the patients with preoperative valgus deformity, the change in MPTA was significant for a greater correction attained in Group P compared with Group E. (5.62 degrees , p = 0.010).

Table III.

Mechanical axis analysis

	All	Group S	Group E	Group P
Preop MAZ	2.61	2.45	2.67	2.7
Postop MAZ	1.97	1.89	2.12	1.9
Change	0.64	0.56	0.55	0.8

MAZ: Mechanical Axis Zone, Group S: Staples, Group E: Titanium eight-plate, Group P: Stainless steel Pedi-plate

We also measured the rate of correction of MPTA, LDFA, and MTFA as degrees per month and compared that between different groups using ANOVA followed by post hoc analysis. This analysis showed significantly faster correction of MPTA in Group P compared with Group S (p = 0.007). This was not seen between Groups E and S (p = 0.8). Correction of LDFA and MTFA was also significantly faster in Group P than S, with p values of 0.001 and 0.007, respectively. The difference between other groups failed to reach statistical significance.

Complications

A list of complications are shown in Tables IV. Overall, complications occurred in 21 of the 77 patients in our analysis (27.3%). Complications included any documented problems including loosening, breakage, pain, knee stiffness, worsening deformity, and inadequate correction. In three cases, the complication necessitated an additional operative intervention.

Table IV.

Complications

	All	Group S	Group E	Group P
Complications # Patients (%)	21/77 (27.3%)	9/18* (50%)	7/24# (29.2%)	5/43*(11.6%)
Most common reasons (# patients)		-Backing out (5) -Broken, Painful, and Inadequate Reduction (2 each)	-Broken Screw (4) -Screw backing out, lost to follow-up, inadequate reduction (1 each)	-Broken screw (2) -Knee stiffness, screw backing out, inadequate reduction (1 each)

^*statistically significant with p:0.000 (between groups S and P)

^#statistically significant with p:0.002 (between groups P and E)

Group S: Staple, Group E: Titanium eight-plate, Group P: Stainless steel Pedi-plate

There was a statistically significant difference in complication rate between Group P and both Group S (p < 0.001) and Group E (p = 0.002). There was no statistically significant difference between Group S and Group E (p = 0.35). In the plate groups, all broken screws were observed in the metaphysis, a phenomenon that has been described previously⁸. Complications mostly occurred in patients with abnormal physes, with 95% of patients that developed a complication having a diagnosis of Blount’s disease, skeletal dysplasia, fibular hemimelia, mucopolysaccharidosis, or hypophosphatemic rickets.

Discussion

Our results support the hypothesis that newer implants and evolving clinical practices have decreased the complication rate over time for patients undergoing angular deformity correction about the knee. The comparison of angular correction did find that all three devices provided angular correction, with some notable differences. The expectation would be that the two different tension-band plate constructs used (Pedi-plate and eight-plate) would behave similarly with respect to angular correction. In fact, our data indicate that the eight-plate did not perform as well as the Pedi-plate or the staple. In 2009, Wiemann et al. found a similar amount of angular correction when comparing Staples (9.9°) to the eight-plate (11.1°)7. We also found that the angular correction occurred more quickly in the staple group (4.16°/year) and Pedi-plate group (4.34°/year) than the eight-plate group (2.09°/year). Stainless steel plates (Group P) showed significantly faster correction of MPTA, LDFA, and MTFA compared to staples which was not seen with the titanium plates (Group E). In 2007, Stevens4 cited a 30% faster correction rate with the eight-plate compared with staples, which was not the case with our population.

Our complication rate, at 27%, was somewhat higher than other series, even accounting for complications that did not lead to further surgical intervention. Previous work has shown 8-17% rate of complications requiring further surgery during guided growth treatment⁵,⁷,¹⁰. Our data indicates a statistically lower complication rate in patients treated with the Pedi-plate than both the eightplate and staple groups. Wiemann reported a similar relationship when he compared eight-plates to staples⁷.

The reasons behind the different effectiveness of the tension band constructs may be related to the implant material type. Both the eight-plate and the Pedi-plate designs are very similar with cannulated screws which are placed into a 2-hole plate. Thus, the stainless steel material must convey increased strength for angular correction. This may be related to the relative pressure felt by the physis using this material; however, this is only conjecture on our part. In addition, most of the implant failures occurred at the junction of the head and neck of the screw, the portion of the screw that is the weakest. The stainless steel may increase the relative shear strength of that implant as we have not seen any hardware failures with the Pedi-plate. It must be noted that Orthofix has recently released a stainless steel version of the eight plates. We have only limited experience with this implant.

Analysis of our angular correction and complication rates based strictly on implant used is useful and allows for the broad generalizations noted above; however, population factors and evolving surgical practices may also have played a role in these results.

One weakness of this study is the difference in diagnoses between groups. While our study population was fairly homogeneous with respect to age, BMI, preoperative deformity, and number of physes treated, there were differences in etiology of angular deformity between our groups, with 40% of the patients in the eight-plate group having a diagnosis of Blount’s disease. However, this was not significantly more than the 31.9% incidence of Blount’s in the staple group. There have been several studies that showed less angular correction in patients with a diagnosis of Blount’s disease⁸,¹⁰,¹¹. This difference in correction could account for some of the difference in angular correction seen between the eight-plate and the Pedi-plate devices, as only 14% of the patients in the Pedi-plate population had Blount’s disease. This same discrepancy may have played into the complication rates as well, as 9 of the 21 patients that had a complication had a diagnosis of Blount’s disease. Additionally, the rate of correction could also depend on the rate of growth at the time of implantation. This rate varies between the boys and girls with the same chronological age and depends more on the level of skeletal maturity. We did not collect skeletal age and did not separate boys and girls in our data analysis. Although the chronological age was not different between our groups, the skeletal age could have been different and may have played a role in the differences seen. The patients in the Pedi-plate group have undergone surgery more recently than the other two groups. The surgeons’ experience with this procedure may have also played a role in our results.

From our data, we conclude that tension band constructs have a significantly lower complication rate than staples used for hemi-epiphyseodesis around the knee. Moreover, this study demonstrated advantages in both angular correction and decreased complications of the stainless steel implant, the Pedi-plate, which is most likely due to implant material properties. Confounding these results was a difference in the number of Blount’s disease patients in each group, the lack of control for skeletal age at the time of implantation, and the possible effect of surgeon experience. Further study, in the form of a prospective randomized trial, would more precisely clarify the differences in performance between these implants.