Retrospective Comparison of Minimally Invasive Plate Osteosynthesis and Open Reduction and Internal Fixation of Radius-Ulna Fractures in Dogs

Abstract

Objective

To compare the efficacy of reduction, time to union, and clinical outcome of radius and ulna fractures stabilized using either minimally invasive plate osteosynthesis (MIPO) or open reduction and internal fixation (ORIF).

Study Design

Retrospective, case-controlled study.

Animals

Dogs with radius and ulna fractures stabilized with plates applied using MIPO (n = 15) or ORIF (n = 15).

Methods

Dogs in each stabilization group were matched for type and location of fracture, age, and body weight. Outcome measures including surgical time, fracture alignment, gap width, plate length, plate bridging and span ratio, working length and screw density, and time to union were compared between the groups using an unpaired t-test. Statistical significance was set at P < .05.

Results

All fractures obtained radiographic union although infection developed in 1 dog in each stabilization group. Dogs treated with MIPO had a significant longer plate working length and lesser screw-density (P < .05). No statistical difference was found in operating time, postoperative alignment, gap width, or time to union (MIPO: 51.9 ± 18.4 days; ORIF: 49.5 ± 26.5 days).

Conclusions

Radius and ulna fractures managed with MIPO had similar alignment, reduction, and time to union as fractures managed with ORIF. Future prospective clinical studies are warranted and should assess healing more frequently and in a standardized manner to compare MIPO to ORIF in a larger population of dogs.

Radius and ulna fractures are common injuries, representing ∼8.5–17% of all fractures incurred by dogs.[1] Toy and miniature breed dogs have a propensity for distal radial and ulnar fractures, which are often the result of minor trauma.[2] Unique morphologic, densitometric, and mechanical differences may predispose smaller dogs to radius and ulnar fractures.[2-4] Complications such as delayed and nonunion as well as refracture after plate removal are reported sequela after stabilization of radius and ulna fractures.[5, 6] Poor intraosseous vascularity and limited periosseous soft tissue coverage may predispose small breed dogs to healing complications.[4] Biologic fracture fixation techniques that limit iatrogenic surgical trauma while yielding appropriate construct stability would appear to be advantageous for facilitating healing of these fractures.

Plating techniques for stabilizing long bone fractures continue to evolve. Emphasis has shifted from obtaining anatomic reconstruction and absolute stability using open reduction techniques to achieving anatomic alignment and appropriate stability using reduction techniques that minimize osseous and soft tissue trauma associated with surgical intervention.[7] Traditional methods of open reduction and anatomic reconstruction require wide exposure of the fracture site, which often results in substantial soft tissue trauma in the region of the fracture and can devascularize fracture fragments. Disruption of the extraosseous blood supply can prolong fracture healing and predispose to infection.[8, 9] In people and more recently in dogs, newer less invasive plating techniques have been developed to decrease the incidence of these complications and improve functional outcome.[10-14] Minimally invasive plate osteosynthesis (MIPO) involves making small plate insertion incisions remote to the fracture and developing an epiperiosteal tunnel between these incisions. The plate is inserted through the epiperiosteal tunnel and fixed to the bone with screws, which are inserted through the plate insertion incisions or through additional stab incisions.[15] The MIPO technique decreases iatrogenic periosteal vascular disruption associated with plate application [16] and may be advantageous when plating radius and ulna fractures in toy breed dogs.

We designed this retrospective, case-controlled study to evaluate the efficacy of the MIPO technique for stabilization of radius and ulna fractures in dogs. Results were compared to a control cohort in which plates were applied by open reduction and internal fixation (ORIF). We hypothesized that fractures stabilized using MIPO technique would achieve radiographic union faster and with fewer complications than fractures stabilized with ORIF. To test our hypothesis, we compared fracture reduction and alignment, characteristics of the implants, time to radiographic union, and complications in dogs managed with MIPO to dogs treated with ORIF matched by age, body weight, and fracture location and configuration.

MATERIAL AND METHODS

Study Design

Medical records (December 2006–May 2011) and radiographs of consecutive dogs with radius and ulna fractures stabilized with MIPO were retrieved. Dogs with incomplete follow-up evaluation were excluded from the study. Information retrieved included signalment, body weight, fracture etiology, concurrent musculoskeletal injuries, and the time interval between injury and surgery. Three dogs managed with MIPO were also included in a separate study, which assessed fracture healing with ultrasonographic evaluation.

For each dog managed using the MIPO technique, a control cohort was established by reviewing our medical record database. This control cohort was chosen by selecting a dog with a radius and ulna fracture stabilized with a plate applied by ORIF and having a similar age, body weight, and fracture configuration[17] and location. Case selection was accomplished by choosing the best possible matching case from a pool of achieved cases with complete follow-up information.

Surgical Procedure

ORIF was performed with the dog positioned in lateral recumbency and the fractured limb nondependent, using dorsomedial or dorsolateral approaches [18, 19] with or without bone grafting according to surgeon preference.

Dogs in the MIPO group were positioned in dorsal recumbency with a foam pad placed under both shoulders. The limb was aseptically prepared using a hanging limb technique [20] and draped in standard fashion for thoracic limb surgery. A sterile cover was applied to the mobile fluoroscopic unit (Siremobil Compact Fluoroscope; Siemens, Iselin, NJ). Orthogonal images of the entire antebrachium, including the elbow and the carpus, were obtained intraoperatively using the fluoroscopic unit to evaluate displacement of the fracture segments and develop a strategy for reduction.

A 2-ring circular fixator (IMEX Veterinary Inc., Longview, TX) was applied to the antebrachium and used to distract the fracture segments and to facilitate and maintain reduction during plate application (Fig 1). The initial fixation wire was placed in the medial-to-lateral plane through the distal radial fracture segment, adjacent, and parallel to the distal articular surface of the radius. In dogs with distal metaphyseal fracture, the wire was placed approximately 10 mm proximal to the radiocarpal joint to allow placement of a screw both proximal and distal to the wire. The wire was attached to the distal ring of the preassembled fixator using wire fixation bolts and nuts. A second fixation wire was then inserted through the proximal radial metaphysis, parallel to the proximal articular surface in the medial-to-lateral plane and attached to the proximal ring in a similar fashion to the first fixation wire. Once the fracture was distracted to anatomic length, reduction was performed as described by Anderson et al.[21] Rotational alignment was evaluated by observing the plane of flexion and extension of the carpus relative to that of the ipsilateral elbow. Rotational malalignment, if present, was corrected by shifting the position of the distal wire about the circumference on the distal ring. The wire was shifted from the cannulation to the slotted portion of the fixation bolts for small adjustments or to different holes on the rings to make larger adjustments. Translational malalignment was improved by sliding the distal fracture segment along the distal wire, which sometimes required slight overdistraction of the fracture. Reduction was assessed during surgery by palpation, visual assessment of limb alignment, and fluoroscopy.

Figure 1. Illustration of the 2-ring circular fixator used to assist in fracture distraction and reduction.

After the fracture segments had been aligned, proximal, and distal insertion incisions were created using a MIPO approach to the radius.[22] After identifying the radiocarpal joint by palpation, a 2-to 4-cm dorsal skin incision was made centered over the distal radial metaphysis. Then, the deep antebrachial fascia was incised between the tendon of the extensor carpi radialis and the tendon of the common digital extensor muscle. A periosteal elevator was used to initiate development of an epiperiosteal tunnel from the distal incision. Metzenbaum scissors were used to create an epiperiosteal tunnel from distal-to-proximal (Fig 2A). A proximal insertion incision was made after measuring and marking the proximal extent of the plate on the skin. The deep antebrachial fascia was incised between the extensor carpi radialis and the pronator teres muscles. The plate was inserted from the proximal insertional incision if the distal fracture segment was caudally displaced and from the distal insertion incision if the distal fracture segment was cranially displaced (Fig 2B).

Figure 2. Application of the fixator and subsequent dissection to develop the epiperiostel tunnel (A) and plate application (B). The fixator consists of two 84-mm diameter incomplete rings that are connected by 2 segments of threaded rod. Two Kirschner wires, 1 inserted through the proximal and 1 through the distal radial fracture segment in the mediolateral plane, are attached to the incomplete rings with cannulated/slotted fixation bolts. Nuts on the inner surface of each ring were sequentially tightened to distract and reduce the fractured radius. Note that the Kirschner wires have deformed as a result of the tension developed during distraction.

After plate insertion, the fluoroscopic unit (Siremobil Compact Fluoroscope; Siemens, Iselin, NJ) was used to acquire orthogonal images of the antebrachium and evaluate the position of the plate. The plate was appropriately positioned over the distal radial segment first and a screw was inserted in the distal screw hole in the plate. The position of the plate over the proximal radial segment was then confirmed and a screw was placed in the proximal end of the plate. Two to 4 screws were inserted in both the proximal and distal radial fracture segments. The number of screws inserted distal to the fracture depended on the length of the distal bone segment. Two screws were placed in the most proximal holes of the plate. Additional screws were inserted in the proximal segment if the bone needed to be reduced against the plate.

Postoperative Care

Postoperative coaptation was not used in either stabilization group. Passive range of motion of the elbow and carpus and cold compression of the surgical area were initiated on the first day after surgery. After discharge, owners were encouraged to continue passive range of motion and cold compression for the first 2 weeks. Nonsteroidal anti-inflammatory drugs were prescribed for 2 weeks. Dogs were allowed to go outdoors for walks on a leash for 10–15 minutes duration several times a day for the first 4 weeks. After the initial recheck, the duration of the walks was increased to 20–30 minutes. Owners were advised to confine their dogs to a cage when unsupervised until their dog's fracture had obtained radiographic union.

Outcome Measures

Orthogonal radiographs of the radius and ulna were obtained pre- and postoperatively and at each scheduled recheck. Fracture reduction and limb alignment were measured in mm (gap width, cortical translation in the frontal plane) and degrees (varus-valgus angulation) using an image-viewing software (Kodak Directview 5.2, Carestream Health, Rochester, NY) on the immediate postoperative images. Rotational alignment was assessed to be normal or mildly or severely misaligned based on subjective assessment of the plane of flexion and extension of the ipsilateral carpus and elbow. Dogs were evaluated 3–4 weeks after surgery, and monthly until the fracture healed. Fracture healing was defined according to radiographic variable (bridging bone seen on both lateral and craniocaudal radiographic projections) and according to clinical findings (normal weight bearing without apparent pain).

Plate type and length (mm) as well as the number of plate holes filled with screws used varied among dogs. The number of screws holes in each dog's plate was recorded. We calculated the plate-bridging ratio as the quotient of the plate length over the radius length (Fig 3). The plate span ratio was calculated by taking the quotient of the plate length over the overall fracture length (Fig 3). Plate-screw density was calculated as the quotient formed by the number of screws inserted over the total number of screw holes in the plate. Working length was defined as the distance between the proximal and distal screws closest to the fracture (Fig 3). Any complications that occurred during the convalescent period were noted. Limb function was evaluated at each recheck with gait and orthopedic examination performed by a board-certified surgeon.

Figure 3. Illustration of a radius with a 13-hole bone plate fixed to the cranial surface of the bone with 7 screws showing the measurements used to calculate the plate-bridging ratio, the plate span ratio, and the working length. Plate-bridging ratio = B/A; plate span ratio = B/D; working length = C.

Data Analysis

All data were reported as mean ± SD. A D'Agostino test of normality was used to test if the difference between pairs followed a Gaussian distribution. An unpaired t-test was used to assess differences between the ORIF and MIPO groups for all outcome measures. Significance was set at P < .05.

RESULTS

Fifteen matched pairs of dogs were evaluated (Tables 1 and 2). No intraoperative complications occurred in either group. All fractures managed with MIPO were reduced indirectly using the fixator, which was considered to be easy to apply and to handle. All fractures managed by ORIF were reduced with direct manipulation of the fracture segments using bone reduction forceps. Corticocancellous allograft (n = 5) or autogenous cancellous graft (n = 2) was used in 7 dogs managed with ORIF.

Table 1. Preoperative Clinical Variables of Matched Cohort Dogs with Radius and Ulna Fractures Stabilized with Minimally Invasive Plate Osteosynthesis (MIPO) or Open Reduction and Internal Fixation (ORIF)

Matched cohorts	ORIF group			MIPO group
	Age (months)	Body weight (kg)	Fracture pattern[35]	Age (months)	Body weight (kg)	Fracture pattern
1	5	2.9	A2	5	3.4	A2
2	8	2.3	A2	6	3.3	A2
3	7	5.6	A2	5	5.2	A2
4	24	10.2	A2	24	6.4	A2
5	28	5.4	A2	36	5.2	A2
6	6	1.6	A2	6	3.2	A2
7	6	4.3	A2	4	5.0	A2
8	36	30.0	A2	21	15.0	A2
9	80	4.0	A2	75	6.5	A2
10	14	2.2	B3	21	2.7	B3
11	8	2.0	A2	24	3.6	A2
12	18	32.7	B2	16	30.0	B2
13	30	40.1	B2	48	18.7	B2
14	144	13.8	B2	115	16.4	B2
15	53	1.7	A2	48	7.2	A2
Mean ± SD	31 ± 38	10.6 ± 12.8		30 ± 31	8.8 ± 7.8

Table 2. Mean ± SD for Outcome Measures (Fracture Reduction, Implant Associated Parameters, and Time to Union) for Dogs in the Minimally Invasive Plate Osteosynthesis (MIPO) and Open Reduction and Internal Fixation (ORIF) Groups. Bolded P-Values Indicate Significant Differences as Determined by an Unpaired t-test

Outcome measures	ORIF group	MIPO group	P- value	Power
Time of surgery (minutes)	113.5 ± 37.9	103.2 ± 40.3	.48	0.1
Radiographic time to union (radius)	49.5 ± 26.5	51.9 ± 18.4	.77	0.06
Fracture gap (mm)	1.4 ± 3.6	3.4 ± 2.4	.29	0.4
Mediolateral translation (mm)	0.4 ± 1.0	1.9 ± 1.3	P < .001
Valgus angulation (degree)	3.7 ± 2.9	4.7 ± 3.4	0.4	0.1
Plate length (number of holes)	7.8 ± 1.9	10.4 ± 2.3	P < .001
Plate span ratio	46.6 ± 25.4	54.4 ± 33.3	0.47	0.1
Plate bridging ratio	0.5 ± 0.1	0.7 ± 0.1	P < .001
Screw density	0.8 ± 0.2	0.5 ± 0.1	P < .001
Working length (number of holes)	1.2 ± 1.9	4.3 ± 1.3	P < .001

A variety of bone plates were used for fracture fixation. Implant failure did not occur in either group; however, plate removal was required in 4 fractures managed with MIPO technique and 3 fractures managed with ORIF. Infection necessitated implant removal in 1 fracture managed with MIPO and 1 fracture managed by ORIF at 9 and 6 months postoperatively, respectively. Both dogs developed a moderate weight bearing lameness. Empirical antibiotic therapy was initially attempted in these dogs; however, lameness only resolved after plate removal. Plates were removed 5–9 months postoperatively in 5 other dogs because pain could be elicited when direct pressure was applied to the plated radius. Owners reported only mild intermittent lameness in these dogs, but both anti-inflammatory and antibiotic therapy failed to resolve the lameness. All of these dogs had ulnar remodeling associated with one of the proximal screws on radiographs suggestive of ulnar impingement. Microbial cultures were negative in these dogs and lameness and pain resolved after plate removal.

The number of screws and their position differed significantly between stabilization groups. Locking implants were used in 1 ORIF dog. MIPO dogs had a significantly greater plate bridging ratio and lesser screw density indicating that most fractures in the MIPO group were stabilized with longer plates and had fewer screws placed than the fractures in the ORIF group. The working length of plates applied in the MIPO group was significantly longer than in the ORIF group (Table 2).

Postoperative fracture alignment was considered satisfactory in all dogs (Figs 4, 5). In the MIPO group fracture segment translation in the frontal plane was significantly greater than in the ORIF group; however, the degree of translation was not considered to be clinically detrimental in any of these dogs. Mean width of the gap in the MIPO group was significantly greater than in the ORIF group. Rotational malalignment was not observed in any fracture.

Figure 4. Preoperative, postoperative, and recheck radiographs of 2 small breed dogs treated with minimally invasive plate osteosynthesis (MIPO) (A–F) and open reduction and internal fixation (ORIF) (G–L). The 2 dogs were matched for type and location of fracture, age, and body weight.

Figure 5. Preoperative, postoperative, and recheck radiographs of 2 large breed dogs treated with minimally invasive plate osteosynthesis (MIPO) (A–F) and open reduction and internal fixation (ORIF) (G–L). The 2 dogs were matched for type and location of fracture, age, and body weight.

Time on follow-up evaluations was variable among dogs, but all dogs were evaluated within 6 weeks after surgery and then every 4–8 weeks until the radial fracture healed. Limb function was considered normal in all dogs based on clinical evaluation at the last follow-up evaluation. All of the radial fractures healed in less than 3 months after the initial surgical procedure, except for 1 ORIF dog, which healed in 130 days. There was no significant difference in time to union between stabilization groups. Ulna fractures had obtained union at the time of the last radiographic evaluation in only 5 ORIF dogs and in 14 MIPO dogs. Although not compared statistically, fractures managed with MIPO subjectively healed with greater callus formation than fractures managed with ORIF.

DISCUSSION

Radius and ulna fractures are typically plated following open reduction.[18] We found the MIPO technique resulted in acceptable reduction and alignment of both simple and comminuted fractures of the radius and ulna. Obtaining proper alignment can be technically demanding during MIPO because the fracture is not exposed and indirect reduction techniques must be used.[20] In our study, both plating techniques resulted in similar frontal plane alignment. In the MIPO group, varus-valgus angulation could be addressed by distracting the fixator ring laterally to correct valgus and medially to correct varus angulation. Placement of nonlocking screws was used to draw the bone to the plate allowing us to align the fracture in the sagittal plane. Although bone segments could easily be slid along the fixator wires, translational malalignment was less effectively corrected than the other types of malalignment. In the MIPO group, 8 fractures had more than 1 mm translation in contrast to the ORIF group in which only 1 fracture had a translational malalignment of more than1 mm. Although not directly compared, simple fractures were more likely to have translational malalignment than comminuted fractures. In people, fracture malalignment is a frequent sequela of MIPO.[13, 23] In a review of femoral and tibial fractures managed with MIPO the incidence of rotational malalignment was 38.5% and 50%, respectively.[23] We did not observe any rotational or frontal angulation malalignment after MIPO. Translational malalignment could only be detected radiographically and did not affect functional outcomes. Another reported technique for performing indirect reduction of radius and ulna fractures when performing MIPO involves placement of an intramedullary pin in the ulna[24]; however, most of the dog's ulnae in our study were too small to allow insertion of an intramedullary pin.

The surgical approach and technique for implant application by the MIPO technique was considered straightforward once reduction had been accomplished. We found the space between the tendons of the extensor muscles to be a simple and effective landmark for inserting the plate distally. One of the differences between this MIPO approach compared to MIPO approaches to the femur and tibia[22] is that the plate insertion tunnel is not entirely epiperiosteal. In a recent cadaveric study, the distal aspect of the fascia of the extensor carpi radialis muscle was observed to be in intimate contact with the periosteum overlying the cranial surface of the radius.[16] Based on our experience, the tunnel may need to be started distally using a periosteal elevator to undermine the aponeurosis between the extensor carpi radialis and the periosteum. One disadvantage of our MIPO technique for radius and ulna fractures is the need to use a temporary external fixator to facilitate indirect reduction of the fracture. Although this additional step requires time and potentially adds cost to the procedure, it did not cause morbidity in the dogs managed with MIPO.

A previous study comparing the pattern of bone healing between fractures treated by ORIF or by MIPO found that dogs treated with MIPO healed with abundant callus formation.[25] In our series, all fractures treated by MIPO healed by secondary bone healing, as suggested by the periosteal callus seen at initial radiographic rechecks. In the fractures stabilized using ORIF, callus was either not seen, or was only apparent 8 weeks after surgery. The observed difference in the pattern of fracture healing may be dependent on both mechanical and biologic factors affecting the fracture.[25] Callus formation is the hallmark of secondary bone healing, which occurs under conditions of relative stability of the implant-bone construct.[27] The callus noted in the fractures stabilized using the MIPO technique in our study suggests that this method provides both an appropriate vascular and mechanical environment to stimulate external callus formation.[26] In a recent cadaveric study evaluating the effect of MIPO on periosteal vasculature of the radius, significantly less vascular disruption occurred in radii stabilized using a MIPO technique compared with radii stabilized using open plating.[16] The periosteal arteries forming the extraosseous blood supply, contribute substantially during the early phase of fracture healing by providing cells, oxygen, and other nutrients to the site of repair.[28] MIPO may facilitate bone healing by preserving the soft tissue envelope and promoting high oxygen tension in the fracture hematoma. In contrast, iatrogenic trauma associated with ORIF may delay the initial phase of fracture healing.[28]

Time to union reported in our study for both MIPO and ORIF is shorter than the median time to union of 10.5 weeks previously reported in dogs that had MIPO stabilization of radial fractures after insertion of an intramedullary pin in the ulna.[24] In the current study, mean time to union for dogs managed with either MIPO or ORIF was less than 6 weeks, but no difference was found between stabilization techniques. Our results would suggest that the variables controlled between the cohorts may influence healing more than the surgical technique itself. Direct comparison of MIPO and ORIF was recently reported in a clinical study evaluating the management of tibial fractures in people with no significant difference identified in healing time or complications.[14] A commentary on this study pointed out several design limitations such as lack of matching of the cases among groups and poor follow-up evaluations, suggesting that the results of this study should be interpreted cautionsly.[30] Our study had similar limitations. Inconsistencies in short-term follow-up evaluations may have led to an overestimation of the healing time, particularly in the MIPO group. Although impossible to demonstrate in this study, several MIPO dogs may have healed well before the first recheck evaluation because the fracture callus was smooth and remodeled at 4 weeks.

One fracture in the MIPO group developed a methicillin resistant Staphylococcus aureus infection in the early postoperative convalescent period. The susceptibility to local infection associated with MIPO has been evaluated both experimentally in rabbits as well as clinical studies in people.[31, 32] In rabbits, the incidence of infection after MIPO was lower than after ORIF, although the difference between groups was not significant.[31] A reported risk factor for a fracture becoming an infected nonunion is extensive periosteal disruption.[32] Devascularized tissue, particularly bone, is an ideal matrix for bacterial growth.[32] Minimally invasive techniques such as MIPO limit the extent of iatrogenic trauma associated with the surgical reduction and stabilization of fractures.[34] One negative attribute of the MIPO technique is that the plate is inserted through small skin incisions and the implant may come in contact with the skin during insertion, possibly inoculating the fracture site.[32, 34]

Major limitations are inherent to our study's retrospective nature. Inconsistent and sporadic follow-up evaluations did not allow us to precisely determine the earliest time to union. The interval between surgery and the first recheck radiographic evaluation varied from 2–8 weeks. The erratic timing of our follow-up radiographic evaluations prevented us from defining precise times to union. More frequent and consistent radiographic evaluations are needed to accurately define the time to radiographic union and have been recommended for dogs managed with MIPO.[25] Another consideration regarding evaluating the time to radiographic union is the accuracy of radiographs in detecting bone healing. In simple fractures managed with ORIF, fractures were anatomically reduced and the plate was applied in a compressive mode. These fractures likely healed by primary bone healing making it difficult to define the time to radiographic union. Time of fracture healing may have been overestimated in dogs with simple fractures in the ORIF group, and underestimated in the fractures managed by MIPO, as secondary bone healing is poorly evaluated using radiographs, whereas primary bone healing is generally overestimated using radiographs.[35]

A further limitation is our lack of consistent and objective measurement of functional outcomes. Although all dogs had returned to normal activity without apparent lameness by the time of the final evaluation, specifics regarding limb function were frequently not included in the medical record. It is our impression that dogs treated with minimally invasive fracture stabilization techniques had a more rapid normalization of limb function than dogs treated with open techniques in this study; this potential advantage of MIPO needs to be evaluated in a prospective study with appropriate outcome measures.

From the results of this preliminary study, we can conclude that the MIPO technique for radius and ulna fractures yields clinical results comparable to those obtained with traditional open plating techniques. We do not, however, have sufficient evidence to advocate MIPO over ORIF for the stabilization of radius and ulna fractures. MIPO of radius and ulna fractures can be technically demanding, and our technique requires temporary placement of an external fixator and intraoperative imaging to obtain precise fracture reduction. Many of the limitations of the current retrospective study could be addressed in a well-designed prospective clinical study. Functional outcome after MIPO and ORIF should be considered to make a conclusive recommendation regarding surgical technique for fracture fixation. Furthermore, more frequent, consistent follow-up recheck evaluations may allow identification of differences in the time to union that this study could not demonstrate.