Comparison of AO Titanium Locking Plate and Screw Fixation versus Anterograde Intramedullary Fixation for Isolated Unstable Metacarpal and Phalangeal Fractures

Abstract

Objective

This study aimed to compare the clinical and radiologic outcomes of AO titanium locking plate and screw (ATLPS) and anterograde intramedullary (AIM) fixation for treating unstable metacarpal and phalangeal fractures.

Methods

Adult patients with isolated fresh unstable metacarpal and phalangeal fractures who met the inclusion criteria were enrolled into this prospective study from July 2013. Patients were divided into ATLPS or AIM groups when they were admitted to our department after considering their work requirement, fracture complexity, and surgeon's experience and were then treated accordingly. Relevant demographic, clinical and preoperative clinical data were collected and analyzed. Clinical examination and radiograph evaluation were performed 1 week and 1, 3, and 6 months postoperatively. Outcome measures were visual analog scale (VAS) scores for pain, total range of motion (ROM) of the injured digit, Quick Disabilities of the Arm, Shoulder, and Hand scores (Quick‐DASH) and grip strength (percentage of the contralateral corresponding digit).

Results

From July 2013 to September 2014, 76 patients were treated by AIM and 71 by ATLPS. Age, sex, time from injury to operation, dominant hand, injury mechanism, fracture location, fracture type and participant occupation were similar in both groups (P > 0.05). Operations were all performed well and followed by uneventful postoperative functional recoveries. At 3‐month follow‐up, all clinical outcomes were significantly better in the AIM than ATLPS group (P < 0.05) except for VAS pain scores. However, at 6‐month follow‐up, the differences were no longer significant, indicating similar results for both types of fixation. Patients in the AIM group developed significantly more complications (P = 0.037). Sick leave was significantly longer in the AIM group (P = 0.02).

Conclusions

AIM outperforms ATLPS in the treatment of unstable metacarpal and phalangeal fractures in the early, but not the later, postoperative period; the latter is associated with significantly more complications. Patients treated by ATLPS require shorter sick leave, which is of particular benefit to workers with specialized manual skills.

Introduction

Metacarpal and phalangeal fractures account for 17.2%–21.3% of all fractures in adults and >30% of all hand injuries1, 2. In most cases, favorable clinical outcomes can be obtained with conservative treatment. However, treatment of markedly displaced, irreducible and rotational fractures can be challenging for surgeons and can lead to impaired function and restricted range of motion (ROM). Surgery has increasingly used to treat unstable hand fractures, being especially indicated when there is shortening of the metacarpus by more than 3 mm or when severe apex‐dorsal angulation is present3.

Intramedullary fixation has been a classical procedure for treating metacarpal and phalangeal fractures; various techniques and types of instrumentation have been proposed to improve postoperative functional recovery and ROM. These include various forms of Kirschner‐wire (K‐wire) pinning (antegrade intramedullary K‐wire, retrograde intramedullary K‐wire, transverse pinning with K‐wire, retrograde cross pinning with K‐wire)4, 5, 6, extramedullary fixation with plates and screws7, 8 and external fixation9. Anterograde intramedullary (AIM) fixation, the conventional treatment for metacarpal and phalangeal fractures, has been reported to have many advantages, including minimal soft tissue dissection, smaller skin incision and potentially less tendon irritation. However, these advantages may be outweighed by inferior stability and a greater incidence of complications. In recent years, fixation with Arbeitsgemeinschaft für Osteosynthesefragen (AO) titanium locking plate and screws (ATLPS) has been used to treat unstable metacarpal and phalangeal fractures and has yielded favorable clinical outcomes. In a retrospective study with 44 patients and 56 fractures, Trevisan and co‐workers used AO mini‐plates and screws to treat low‐severity metacarpal and phalangeal fractures and achieved fracture reduction in 94.6% of patients; however, by the final check‐up approximately half the patients had developed one or more complications10. Agarwal and colleagues conducted a prospective study of 20 fractures to identify whether the thickness influences the postoperative clinical outcomes and complication rate and found no differences compared with the widely used Stryker‐Leibinger system11. Omokawa et al. conducted a prospective study to evaluate clinical results of miniature titanium plate for open reduction and internal fixation of unstable metaphyseal fractures of the metacarpal and phalangeal bones12. With 55 patients and a minimum follow‐up of 1 year, 5 patients (9.0%) developed complications, including fracture re‐displacement, collapse or absorption of the condylar head and superficial infection; and the objective outcomes were reasonable.

However, most reports describing these two techniques are non‐comparative descriptive studies or case reports and thus provide insufficient evidence to identify the optimal type of fixation for this injury. The present prospective study was conducted to compare the outcomes of ATLPS and AIM in the treatment of unstable metacarpal and phalangeal fractures. The aims were: firstly, to compare postoperative functional recovery; secondly, to compare the complication rates; and thirdly, to identify patients who could benefit from either type of fixation. The following variables were assessed: patient satisfaction using a visual analog scale (VAS), total ROM of the injured digit, Quick Disabilities of the Arm, Shoulder, and Hand (Quick‐DASH13) scores and grip strength (expressed as a percentage of the strength of the corresponding digit on the contralateral side).

Patients and Methods

This study was approved by the Institutional Review Board of the Third Hospital of Hebei Medical University. All study patients provided written informed consent. From July 2013, consecutive patients with isolated metacarpal and phalangeal fractures treated in our department were initially assessed and divided into AIM or ATLPS groups on admission to our hospital after considering the patient's work requirement, fracture complexity, and surgeon's experience. The following inclusion criteria were used to select patients for operative fixation with either AIM or ATLPS: age >18 years; closed extra‐articular metacarpal or phalangeal fracture; oblique or spiral fracture with displacement of >3 mm or rotational deformity of >5° or marked lateral angulation of >30°; and no history of fracture in the injured hand. Exclusion criteria were as follows: age <18 years with fracture of immature bone; old fracture or operated fractures; open injury or intra‐articular fracture; multiple metacarpal or phalangeal fractures or concurrent injuries at other sites; and refusal to consent to random allocation of type of fixation. The inclusion and exclusion criteria were applied rigorously to maximally eliminate the negative effects of any confounding variable related to the clinical outcomes of either form of fixation. Relevant clinical and demographic data were prospectively collected from both AIM and ATLPS groups and analyzed.

Surgical Techniques

All operations were performed by a three‐surgeon surgical team (led by attending physician Bing Zhang, who was assisted by two residents). In the ATLS group, operations were performed with tourniquet hemostasis under brachial plexus block or local anesthesia and a longitudinal dorsal approach was used. The fracture line was exposed by retracting the extensor tendons bilaterally for metacarpal fractures and by an extensor tendon‐splitting incision for phalangeal fractures. After elevating the periosteum, the fracture site was exposed and the fracture reduced and stabilized with an AO miniplate using appropriately sized screws (1.2–2.0 mm). Postoperative immobilization was not required in patients without tendon or nerve impairment. In patients who had undergone tendon or nerve repair, immobilization was achieved with external fixation with a plaster cast for 3 weeks. Passive motion was started on the second postoperative day and active movement in the third postoperative week. Patients were allowed to resume activities of daily living only after bony union had been observed on follow‐up plain radiographs. Figure 1 shows X‐ray films and photos of a 35‐year‐old, right‐handed, male manual worker who sustained a closed fracture of the distal third of the right fifth metacarpal, with 65° of angulation. At final follow‐up, he had completely recovered and returned to his previous job.

Figure 1.

(A) X‐ray film: a 35‐year‐old right‐handed manual worker sustained a closed right fifth distal‐third metacarpal fracture with an angulation of 65°. (B) Intraoperative view and (C) postoperative X‐ray film of fracture fixation with AO titanium locking plate and screws. (D, E) Normal functional recovery was noted at the final follow‐up.

For patients in the AIM group, an approximately 5 mm long incision was made in about the proximal joint area of the involved metacarpal or phalanx. Using a drill, an entry point for a K‐wire was made at the base of the metacarpal or phalanx. The distal tips of two K‐wires of suitable size (0.8–1.0 mm) were pre‐bent approximately 20° to resemble a hockey stick, and the fracture reduced using a Jahss maneuver14. The wires were then inserted into the intramedullary canal by tapping with a hammer in divergent directions under image intensification. The wires were then cut and the skin closed over the cut ends. Postoperatively, splinting was maintained for approximately 4–6 weeks. After splint removal, patients were encouraged to perform passive and active wrist and finger motions. The wires were removed through a small incision under local anesthesia in an outpatient setting 6–8 weeks, but no more than 10 weeks, postoperatively. Figure 2 shows the X‐ray films and photos of a 29‐year‐old male patient with oblique fractures of the left third phalanx treated with fixation with two K‐wires.

Figure 2.

X‐ray films and photos of a 29‐year‐old male patient with an oblique fracture of the left third phalanx treated with fixation with two K‐wires. (A, B) Preoperative AP and lateral X‐ray films. (C, D) Postoperative AP and lateral X‐ray films.

Clinical Evaluations

Patients underwent follow‐up examinations at 1 week, 6 weeks, 3 months and 6 months postoperatively. Clinical assessment was performed independently by three trained physiotherapists. The mean value of measurements taken by the three physiotherapists was used in the analysis. The variables assessed included total ROM of the finger, grip strength and Quick‐DASH scores. Total ROM of the finger is expressed as the sum of movement range in each joint: metacarpophalangeal (MP) joint (normal, 0°–85°), proximal interphalangeal (PIP) joint (0°–110°) and distal interphalangeal (DIP) joint (0°–65°)15. Using a Jamar Hydraulic Hand Dynamometer (Patterson Medical Holdings, Warrenville, IL, USA)16, the grip strength of the involved finger was measured by having the patient squeeze the handle of the dynamometer maximally. The grip strength of the corresponding finger in the contralateral, uninjured hand was measured and grip strength calculated as a percentage of that of the contralateral hand. The Quick‐DASH questionnaire consists of 30 items; 21 address the ability to perform specific activities and nine address symptoms. DASH disability/symptom score was calculated as ([sum of n responses]− 1) × 25, where n is the number of completed responses. Thus, the range of responses was from 0 to 100, higher scores indicating greater patient disability17. At each follow‐up evaluation, pain during daily activities was assessed using a VAS scale, 0 indicating no pain and 10 the most severe pain.

Complications and, for working patients, length of sick leave, were recorded at all follow‐up examinations. A radiologist read all the radiographs taken pre‐ and postoperatively and at each follow‐up.

Statistical Analysis

Pre‐ and post‐operative variables were evaluated and analyzed based on data format. Because the data in this study were not normally distributed, the Mann–Whitney U test was used to compare the continuous variables, namely age, time from injury to operation, total ROM of the injured finger, Quick‐DASH scores and VAS scores. Comparisons between the groups were analyzed using the χ² or Fisher's exact test for the categorical variables, namely sex distribution, injury mechanism, lateral angulation and grip strength. Statistical significance was set at P < 0.05. Subgroup analyses could not be performed because of the risk of under‐powering the data. SPSS for Windows, Version 16.0 (SPSS, Chicago, IL, USA) was used for analysis and modeling of the data.

Results

Participant Selection and Preoperative Clinical Data

Between July 2013 and September 2014, 373 consecutive metacarpal and phalangeal fractures in 373 patients treated in our department were included for initial assessment. After application of inclusion and exclusion criteria, 226 patients were excluded: 145 because they did not meet inclusion criteria, 45 because they declined to participate and 36 for other reasons. Figure 3 presents the details of participant selection. Thus, 147 patients were included in this study. Seventy‐six participants (71 men, 5 women; median age 34 years [range, 18–56 years]) were allocated to the AIM group and 71 (68 men, 3 women; median age 33 years [range, 18–55 years]) to the ATLPS group. Demographic and preoperative clinical data are provided in Table 1. Age, sex, time from injury to operation, dominant hand, injury mechanism, fracture location, fracture type and participant occupation were similar in the two groups (all P > 0.05). By final follow‐up, radiographic bony union had been achieved in all patients.

Figure 3.

Flow diagram showing the steps in patient selection and group outcomes at 3‐ and 6‐month follow‐ups.

Table 1.

Preoperative baseline characteristics according to type of fixation

Variable	AIM (n = 76)	ATLS (n = 71)	P value
Sex (M/F)	71/5	68/3	0.720
Age (years)	34 (18–56)	33 (18–55)	0.326
Time from injury to operation	2.3 d (2 h to 7 d)	2.6 d (4 h to 8 d)	0.172
Hand (dominant/non‐dominant)	67/9	67/4	0.097
Mechanism (axial blow/other)	27/49	22/49	0.560
Fracture location (cases)			0.957
Proximal third	12	10
Middle third	29	28
Distal third	35	33
Fracture type (number of cases)			0.713
Lateral angulation	36	31
Spiral	17	14
Oblique	23	26
Occupation (manual worker/others)	48/28	44/27	0.882

Postoperative 3‐ and 6‐Month Assessments

At the 3‐month follow‐up, five patients in the AIM group and three in the ATLPS group were lost to follow‐up because they had relocated or could not be reached by telephone. For the remaining patients, all clinical outcome measures except VAS score favored the AIM group (Table 2). Median total ROM of the injured finger was significantly better in the AIM than in the ATLPS group (P = 0.004). Median grip strength was greater in the AIM than the ATLPS group, with an 11% difference between medians (P < 0.01). Median Quick‐DASH scores were significantly smaller (i.e., better) in the AIM than the ATLPS group (difference of 5.8 between medians [P < 0.01]). Patient's satisfaction regarding pain was very high in both groups; the one point difference between medians was not statistically significant (P = 0.721).

Table 2.

Outcome measures according to type of fixation and time after surgery

Variable	Three months postoperatively			Six months postoperatively
	AIM group (n = 71)	ATLS group (n = 68)	P value	AIM group (n = 71)	ATLS group (n = 66)	P value
TROM	231 (130–260)	215 (130–248)	0.004	247 (170–260)	238 (170–260)	0.455
VAS	1 (0–4)	2 (0–4)	0.721	0 (0–2)	0 (0–3)	1
Grip strength value (%)	84 (63–100)	73 (52–98)	<0.001	96 (78–104)	94(76–103)	0.870
Quick‐DASH	6.2 (0–20.4)	12.0 (0–27.8)	<0.001	4.1 (0–13.3)	4.6 (0–14.8)	0.462

TROM, active total range of motion of injured finger. Mann‐Whitney U test; data are presented as mean and range.

Six months postoperatively, data were collected for 71 patients in the AIM group and 66 in the ATLPS group. Seven patients in the AIM and five in the ATLPS group were unable to present for clinical assessment for personal reasons; data on these patients were obtained by telephone interview following our instructions. We found no significant differences between the groups for any variable (Table 2), including total ROM of the injured finger (P = 0.455), median VAS scores (P = 1), grip strength (P = 0.870) and DASH scores (P = 0.462).

Assessment of Postoperative Complication

Complications tended to be more common in the AIM (n = 11) than the ATLPS group (n = 4); however, this difference was not significant (P = 0.466). In the AIM group, three patients experienced slight loss of reduction that did not require revision; three wire migration (one to the skin and two within the fracture site), which were treated with early wire removal; and one malrotation of the operated finger—this patient underwent AIM revision rather than ATLPS fixation. Also in the AIM group, two patients developed superficial infection, which was successfully managed with topical antibacterial agents, and one irritation of the extensor tendon. Finally, one patient in the AIM group had suspected nerve irritation, which was treated by wire removal. Four complications developed in the ATLPS group over the course of follow‐up: two cases of stiffness were resolved with removal of instrumentation and extensor tenolysis and two patients with delayed union underwent surgical revision and had achieved bony union at follow‐up. Sick leave was significantly longer in the AIM than the ATLPS group (P < 0.01).

Discussion

The standard treatment for metacarpal and phalangeal fractures has been AIM fixation; some studies have demonstrated that it has advantages over other surgical methods4, 18, 19. However, few studies have compared outcomes between AIM and ATLPS fixation, presumably because of the relatively recent introduction of ATLPS for treating this injury. We therefore conducted this study to investigate whether these modalities differ with respect to hand function and postoperative complications in cases of isolated, closed, unstable metacarpal and phalangeal fractures.

At 3‐month follow‐up, AIM fixation significantly outperformed ATLPS in terms of total ROM, Quick‐DASH scores and recovery of grip strength; however, these differences were not significant at the 6‐month follow‐up. There were no significant differences in VAS pain scores between groups at either follow‐up point. These results are consistent with those of previous studies. Fujitani et al. compared the outcomes of AIM with those of a low‐profile miniplate for treatment of fifth metacarpal neck fractures and found that total ROM of the fifth finger was better in the AIM group at 3 months but not at 6 months18. In a study by Ozer et al. of treatments for extra‐articular metacarpal fracture, slightly greater total ROM was observed in the AIM group (237°–250°) than in the plate‐and‐screws group (228°–248°); however, this difference was not significant20. The reason for this result is not difficult to explain. Minimally invasive surgery, requiring little soft tissue dissection and innocent nerves and tendons, may have facilitated the rehabilitation process, especially with isolated closed metacarpal and phalangeal fractures, early (up to 3 months postoperatively) but not later (6 months postoperatively) in the recovery period. The AIM group required longer sick leave than the ATLPS group (median of 28 vs. 17 days, respectively), this being partially attributable to splinting in the AIM group. Therefore, ATLPS may be a better alternative for those who require immediate mobility and return to previous work, especially manual workers with specialized skills.

The main complications of K‐wire fixation are migration of the K‐wires and malrotation of the digit, which requires surgical revision or conversion to ATLPS. To prevent these complications, a bouquet configuration of the wires has been suggested by Sletten et al. 21 and Foucher22 for treatment of fifth digit metacarpal neck fractures. Angled cross‐pinning and insertion of as many K‐wires as possible are recommended to prevent rotational instability and migration. Additionally, specific designs have been introduced to resolve these problems, including using stout, modified K‐wires; pins attached to an insertion handle; and a locking device8, 23. There were two cases of finger stiffness in the ATLPS group; both required surgical revision and postoperative rehabilitation. This mobility deficit is attributable not only to the extensive surgical incision, but also to the extensor apparatus and adhesions to the adjacent joint. In their study, Page and Stern encountered extensor tendon rupture and other complications that they attributed not only to the non‐locking design of the plates, but also to the frequent use of plates in open phalangeal fractures24. In addition, we encountered two cases of delayed consolidation, which could potentially have greatly influenced mobility and functional outcomes. These patients underwent timely surgical revision and had achieved bony union at the final follow‐up. No case of malunion requiring revision was encountered in this study, as has been reported several times elsewhere25. We also encountered no head necrosis, neurologic and extensor ligament complications, plate loosening or breakage19, 26. The relatively low incidence of complications may be closely related to the lack of severity of the injuries. As described previously, patients with open or multiple fractures and concomitant injuries were excluded.

Shortcomings and Limitations

Despite its prospective design, this study has some limitations. First, patients were not randomized to one treatment or the other; the type of fixation was selected after considering the patient's work requirement, fracture complexity, and surgeon's experience, which may have lowered test power. However, both groups were homogeneous for most preoperative data (except lateral angulation), which partially made up for this. Second, the duration of follow‐up was relatively short (6 months) and several patients were lost to follow‐up, which may have produced some bias.

Conclusions

In conclusion, this prospective comparison study has demonstrated that in terms of total ROM of the injured finger, Quick‐DASH scores and recovery of grip strength, AIM outperforms ATLPS in the management of unstable metacarpal and phalangeal fractures in the early, but not in the later, postoperative period. However, with respect to postoperative complications and sick leave for rehabilitation, AIM performs worse than ATLPS. Patients treated by ATLPS require less sick leave, which is of particular benefit to workers with specialized manual skills. These results may assist both surgeons and patients in selecting a surgical strategy for treating isolated unstable metacarpal and phalangeal fractures.