Treatment of terrible triad injuries at a mean follow-up of nine years

Abstract

Background

To evaluate the sustainability of the early clinical and radiological outcomes of terrible triad injuries at long-term follow-up.

Methods

Twelve consecutive patients who underwent fixation of terrible triad injuries with minimum of seven years of follow-up. Functional and radiological outcomes at one year and final follow-up were compared. We assessed Mayo Elbow Performance Score, Quick-Disability of the Arm Shoulder and Hand, and modified-American Shoulder and Elbow Surgeons Scores. Radiological evaluation included Broberg and Morrey classification and Hastings classification for heterotopic ossification.

Results

The mean age of patients was 55 years, with a median follow-up of 9.3 years. At final evaluation, mean flexion, extension, supination, and pronation were 145°, 6°, 82°, and 80°, respectively; mean Mayo Elbow Performance Score, modified-American Shoulder and Elbow Surgeons Scores, and Quick-Disability of the Arm Shoulder and Hand scores were, respectively, 97, 92, and 4.9 points. There was no statistical difference between early and final follow-up range of motion, Mayo Elbow Performance Score, and Quick-Disability of the Arm Shoulder and Hand. However, radiological changes were observed in 66% of the patients at final follow-up. Two patients underwent reoperation, with final satisfactory results.

Conclusion

Our results suggested that using a standardized protocol, satisfactory clinical outcomes at early follow-up could be maintained over time. However, early complications and osteoarthritic changes at long-term follow-up can be expected.

Level of evidence

Therapeutic IV Cases series.

Introduction

In 1996, Hotchkiss introduced the term “terrible triad” to describe posterior dislocation of the elbow with associated fractures of the radial head and the coronoid process of the ulna.¹ Historically, this complex injury had poor outcomes with a high complication rate, including stiffness, instability, pain, posttraumatic arthritis, and heterotopic ossification (HO).^2–4

Nowadays, the paradigm of terrible triad injuries (TTIs) prognosis is changing. Recent literature has improved our understanding of elbow anatomy and biomechanics along with the pathoanatomy of this injury, thereby allowing for the development of a systematic approach for its treatment and rehabilitation.⁵ It has been proven that by using standardized surgical and rehabilitation protocols, better results can be achieved.^6–8

To our knowledge, the long-term results of the TTI have been vaguely reported in the literature. A long-term registry of this injury provided the opportunity to study the long-term outcome after surgical treatment to investigate if this type of treatment is durable over time and to see whether the functional results at early evaluation persist or improve.

The purpose of this study was to evaluate the sustainability of the early clinical and radiological outcomes of TTIs at long-term follow-up using a standard treatment protocol that includes coronoid repair, radial head open reduction and internal fixation (ORIF) or replacement, and lateral collateral ligament (LCL) repair at a minimum of seven years after injury.

Methods

A retrospective study was carried out with the approval of our institutional review board.

We included patients with acute TTI treated by a single surgeon operated between 2005 and 2010, using a standardized surgical and rehabilitation protocol, which included coronoid-brachialis complex fixation, radial head fixation or prosthetic replacement, and a repair of the lateral ulnar collateral ligament, with a minimum of seven years of follow-up. We excluded skeletally immature patients, open fractures, and subacute injuries (more than three weeks after original trauma). Fourteen consecutive patients met the inclusion criteria, but two were lost at the final follow-up. Both patients showed satisfactory clinical and radiological outcomes at three and five years, respectively, but they could not be reached for the final evaluation. Finally, 12 patients were included in the study. Eight patients were male and four female. The mean age of the study population at final follow-up was 55 years (range 26–72 years).

Functional assessment

Clinical evaluation was performed by two independent researchers using a handheld goniometer to measure the active range of motion of the elbow in flexion, extension, and forearm rotation. Elbow stability was assessed using the varus-valgus stress test and lateral pivot shift tests. Both subjective and objective clinical outcomes were assessed using Mayo Elbow Performance Score (MEPS), Quick-Disability of the Arm Shoulder and Hand (Q-DASH), modified-American Shoulder and Elbow Surgeons Scores (m-ASES), and Visual Analogue Score (VAS) for pain.^9–11

Functional outcomes between early and final follow-up were compared. The early evaluation took place at a median of 11 months after injury (range 10–18). Mean time between the injury and the index procedure was 6.9 days (range 5–19). The long-term evaluation took place at a median of 9.4 years (range 7–12).

Radiological assessment

Preoperative and postoperative radiographic assessment with standard posteroanterior and lateral X-rays of the elbow was performed by two authors.

We categorized the coronoid fractures according to the Regan and Morrey¹² classification and the radial head fracture according to the Mason–Johnston¹³ classification (Table 1). Radiological evaluation included Broberg and Morrey¹⁴ classification for osteoarthritis (grade 0, no arthritis; grade 1, slight joint space narrowing with minimum osteophyte formation; grade 2, moderate joint space narrowing with moderate osteophyte formation; and grade 3, severe degenerative change with gross destruction of the joint), while HO was classified according to Hastings (class I: radiographically demonstrable elbow or forearm HO without functional limitation; class IIA: limitation in elbow flexion/extension plane; class IIB, limitation in forearm pronation/supination plane; class IIC: limitation in both planes of motion; class III: ankylosis that eliminates either elbow flexion/extension, pronation/supination or both).¹⁵

Table 1.

Early results.

Case	Sex	Age (years)	Time from injury to index procedure (days)	Radial head (Mason)	Coronoid (Morrey)	Radial head treatment	Follow-up (months)	Broberg and Morrey	Grewal	Flexion	Extension deficit	Pronation	Supination	MEPS	Q-DASH
1	M	42	19	3	2	ORIF	18	1	–	140	10	90	90	100	2
2	M	26	7	2	1	ORIF	12	0	–	140	20	90	80	80	7
3	F	69	6	2	2	Partial excision	10	2	–	150	0	90	90	100	7
4	F	48	7	2	1	ORIF	12	0	–	150	10	90	70	100	6
5	F	72	6	3	1	Prosthesis	10	0	None	140	20	90	90	90	15
6	F	63	4	3	2	ORIF	11	1	–	140	10	90	90	100	7
7	F	70	4	3	1	Prosthesis	12	0	None	125	10	70	90	95	13
8	F	58	14	3	1	Prosthesis	11	0	None	140	10	90	90	90	15
9	F	65	10	3	1	Prosthesis	11	0	None	140	0	90	70	85	0
10	F	56	5	3	1	Prosthesis	12	0	None	140	15	90	70	95	15
11	M	38	8	1	1	ORIF	11	0	–	150	0	90	90	100	2
12	M	62	5	2	1	ORIF	10	0	–	140	5	90	80	100	7
Mean		55	7				11			141	9	88	83	94	8
Median		60	6				11			140	10	90	90	97	7
Minimum		26	4				10			125	0	70	70	85	2
Maximum		72	19				18			150	20	90	90	100	15

MEPS: Mayo Elbow Performance Score; ORIF: open reduction and internal fixation; Q-DASH: Quick-Disability of the Arm Shoulder and Hand.

For cases treated by radial head replacement, Grewal¹⁶ classification was analyzed in order to assess periprosthetic lucency.

Operative protocol

All surgeries were performed by the senior author. Following the approach described by Pugh and McKee,¹⁷ we repaired the damaged structures from deep to superficial, as seen from the lateral approach (coronoid to anterior capsule to radial head to lateral ligament complex to common extensor origin).

The lateral approach (Kaplan’s intermuscular septum between extensor carpi radialis brevis and extensor digitorum communis) was used if there was no instability on valgus stress tests; otherwise a posterior global approach was used. The coronoid fractures were addressed first in all patients. If radial head arthroplasty was chosen as treatment, the neck cut was performed to improve the access to the coronoid, and then the coronoid was fixed. Large coronoid fractures were treated with screws, whereas small fragments were treated with suture anchors in repair of the anterior capsule.

Following which, fractures of the radial head were addressed. Generally, the primary goal was to fix the fracture when there was one or two fracture fragments of the head (seven patients). If fracture comminution (three or more fragments), impaction, cartilage damage, or an associated radial neck fracture indicated that a stable anatomic reduction was not feasible, we preferred radial head arthroplasty (Monoblock, nonanatomic titanium implant, Wright Medical Technology, Arlington, TN, USA) (four patients). Minor fragments (<25% of the radial head) that were too small or damaged to fix were debrided, and the residual intact radial head was left in situ (one elbow).

Once the coronoid and radial head were treated, the lateral ligament complex was repaired back with suture anchors placed at the isometric point of the lateral epicondyle. The elbow was then examined under an image intensifier to confirm reduction and to assess stability through a range of motion. For cases with persistent instability or articular incongruence, we explored and repaired the medial collateral ligament (MCL) through a separate medial approach using suture anchors (two cases). It was not necessary to apply a hinged external fixation in any case (Figures 1–4).

Postoperative management

Postoperative rehabilitation protocols were similar in all patients. Patients were initially placed in a posterior plaster splint. The splint was removed seven days after surgery, and active and active-assisted motion was started. Full ROM was allowed with the forearm in pronation or neutral with supination occurring with the elbow at 90° of flexion or greater. Terminal extension in supination was avoided for approximately three weeks. X-rays were performed at 1, 3, 6, 12, 18, and 24 weeks after surgery. The radiological evaluation was subsequently performed every three months in the first year and annually thereafter.

Statistical analysis

Continuous data are presented as the median and range. We evaluated differences in continuous variables by Wilcoxon signed-rank test to compare paired results over time. Two-tailed p values less than .05 were considered statistically significant.

Results

Early results

At a median of 11 months after injury (range 10–18), the median flexion, extension, supination, and pronation were 140° (range 125–150), 10° (range 0–20), 90° (range 70–90), and 90° (range 70–90), respectively. The median DASH was 7 (range 2–15) and the median MEPS was 97.5 (range 80–100). Three patients had signs of arthritis: grade 1 in two elbows and grade 2 in one elbow. No residual elbow instability was observed. No patient showed HO at the early evaluation.

Long-term results

At a median of 9.3 years (range 7–12), the median flexion, extension, supination, and pronation were 145° (range 140–150), 6° (range 0–20), 82.5° (range 70–90), and 80° (range 40–90), respectively; median MEPS, m-ASES, and Q-DASH scores were, respectively, 100 (range 80–100), 93 (range 78–100), and 2 points (range 0–20). The mean VAS was 0.34 (range 0–2).

Regarding the radiographic evaluation (Table 2), eight elbows had signs of arthritis affecting the ulno-humeral joint with associated radio-capitellar compromised in three of them. According to the Broberg and Morrey classification, there was five elbows with grade 1 and three elbows with grade 2, demonstrating significant progression of the posttraumatic changes at the final follow-up (eight patients) compared to the early follow-up (three patients). However, none of the clinical scores showed correlation with the radiographic changes. HO class 1 was observed in six patients and class IIC in one patient, and of the five patients with radial head prosthesis, four of them showed signs of periprosthetic lucencies (three mild and one moderate). At final follow-up there was not a significant difference in terms of HO and range of motion between patients treated with radial head replacement and ORIF.

Table 2.

Long-term results.

Case	Sex	Age (years)	Time from injury to index procedure (days)	Radial head (Mason)	Coronoid (Morrey)	Radial head treatment	Follow-up (months)	Broberg and Morrey	Hasting and Graham	Grewal	Flexion	Extension deficit	Pronation	Supination	MEPS	Q-DASH	m-ASES
1	M	57	19	3	2	ORIF	145	2	I	–	140	5	90	80	100	2	92
2	M	35	7	2	1	ORIF	113	2	IIC	–	145	20	90	80	90	2	94
3	F	78	6	2	2	Partial excision	106	2	I	–	150	5	90	80	80	20	78
4	F	60	7	2	1	ORIF	140	1	I	–	150	0	90	80	100	0	100
5	F	81	6	3	1	Prosthesis	110	0	–	Mild	150	0	90	85	100	2	88
6	F	78	4	3	2	ORIF	140	1	I	–	140	15	90	90	100	7	86
7	F	78	4	3	1	Prosthesis	95	1	–	Mild	145	0	90	90	100	0	100
8	F	66	14	3	1	Prosthesis	127	1	–	Mild	150	5	90	90	100	7	96
9	F	73	10	3	1	Prosthesis	110	0	I	None	140	5	90	75	95	2	92
10	F	63	5	3	1	Prosthesis	124	0	–	Moderate	140	10	90	70	100	15	88
11	M	45	8	1	1	ORIF	90	0	–	–	150	0	90	90	100	0	100
12	M	70	5	2	1	ORIF	85	1	I	–	140	5	90	80	100	2	96
Mean		65	7.9				115.4				145	5	90	82	97	4	92
Median		68	6.5				111.5				145	5	90	80	100	2	93
Minimum		35	4				84				140	0	90	70	80	0	78
Maximum		81	19				180				150	20	90	90	100	20	100

m-ASES: modified-American Shoulder and Elbow Surgeons Scores; MEPS: Mayo Elbow Performance Score; ORIF: open reduction and internal fixation; Q-DASH: Quick-Disability of the Arm Shoulder and Hand.

Early and late results compared

As shown in Table 3, the median arc of elbow flexion increased from 130° (range 115°–150°) to 140° (range 125°–150°), with median flexion improving from 140° (range 125°–150°) to 145° (range 140°–150°) and a median flexion contracture decreasing from 10° (range 0°–20°) to 5° (range 0°–20°). The median arc of forearm rotation decreased from 175° (range 160°–180°) to 160° (range 160°–180°). The median Broberg and Morrey score increased from 0 points (range 0–2 points) to 1 point (range 0–2 points) at long-term follow-up. There was no statistical difference between early and final follow-up range of motion, MEPS, and Q-DASH.

Table 3.

Differences between early and long-term results.

Case	Flexion	Extension	Flexion arc	Pronation	Supination	S/P arc	MEPS	Q-DASH	Broberg and Morrey	Hasting and Graham
1	140/140	10/5	130/135	90/90	90/80	180/170	100/100	2/2	1/2	–/I
2	140/145	20/20	120/125	90/90	80/80	170/170	80/90	7/2	0/2	–/IIC
3	150/150	0/5	150/145	90/90	90/80	180/170	100/80	7/20	2/2	–/I
4	150/150	10/0	140/150	90/90	70/80	160/170	100/100	6/0	0/1	–/I
5	140/150	20/0	120/150	90/90	90/85	180/175	90/100	15/2	0/0	–/–
6	140/140	10/15	130/125	90/90	90/90	180/180	100/100	7/7	1/1	–/I
7	125/145	10/0	115/145	70/90	90/90	160/180	95/100	13/0	0/1	–/–
8	140/150	10/5	130/145	90/90	90/90	180/180	90/100	15/7	0/1	–/–
9	140/140	0/5	140/135	90/90	70/75	160/165	85/95	0/2	0/0	–/I
10	140/140	15/10	125/130	90/90	70/70	160/160	95/100	15/15	0/0	–/–
11	150/150	0/0	150/150	90/90	90/90	180/180	100/100	2/0	0/0	–/–
12	140/140	5/5	135/135	90/90	80/80	170/170	100/100	7/2	0/1	–/I
Mean	141/145	9/5	132/139	88/90	83/82	171/172	94/97	8/4
Median	140/145	10/5	130/140	90/90	90/80	175/170	97.5/100	7/2
Minimum	125/140	0/0	115/125	70/90	70/70	160/160	85/80	0/0
Maximum	150/150	20/20	150/150	90/90	90/90	180/180	100/100	15/20
p value	0.3	0.2		0.9	0.6		0.3	0.1

MEPS: Mayo Elbow Performance Score; Q-DASH: Quick-Disability of the Arm Shoulder and Hand; S/P: Supination-Pronation Arc.

Subsequent procedures and complications

Two patients underwent reoperation. One radial head nonunion was treated with implant removal and fragment excision and one elbow stiffness required a manipulation under anesthesia and both showed final satisfactory results at the final follow-up.

Discussion

A series of patients with TTI of the elbow treated by coronoid-brachialis fixation, radial head replacement, or fixation and lateral ligament complex repair were analyzed in the present study. Despite reoperation was required in 16% of cases, excellent early clinical outcomes achieved in 11 of the 12 patients were maintained at long-term follow-up (median of 9.3 years) with this approach.

Since the original description of the pathology’s treatment outcomes, TTIs were associated with high complication rates and poor clinical prognosis due to stiffness, elbow instability, and chronic pain. To our knowledge, there are only two publications reporting long-term outcomes of TTI’s treatment. Ring et al.² published 11 cases with a mean follow-up of seven years. In these series, the radial head was resected in four patients, the LCL was repaired only in 3 of the 11 patients and none of the coronoid fractures were fixed. This approach led to unsatisfactory results in 7 of the 11 patients at the final follow-up. More recently, Muñoz-Mahamud et al.¹⁸ reported 16 patients at a mean follow-up of eight years. The coronoid process was not fixed in six of the patients and the LCL was not repaired in half of the 16 patients. At final follow-up, the authors reported that 30% of cases described discomfort or pain and 25% described difficulties in daily activities. Conversely, we have systematically repaired the coronoid process and the LCL in our series, which has allowed us to achieve adequate joint stability and lead to satisfactory clinical outcomes at a mean follow-up of nine years.

Figure 1.

Antero-posterior radiograph of the elbow showing a postero-lateral dislocation, as a part of a TTI.

Figure 2.

Lateral radiograph of the elbow, demonstrating the elbow dislocation in association with a radial head and coronoid fracture.

Figure 3.

Antero-posterior radiograph of the elbow at 15-year follow-up.

Figure 4.

Lateral radiograph of the elbow at 15-year follow-up.

Over the last decade several publications have demonstrated significant improvements in clinical outcomes regarding the treatment of TTI compared with classic reports.^4,5,8,19 Despite minor differences, most protocols recommend fixation or arthroplasty of radial head fractures in association with treatment of the fracture of the coronoid process, followed by repair of the LLC through transosseous sutures or the use of anchors. Cases with residual instability have been treated with an articulated external fixator and/or repair of the MCL.^5–8

Recent studies have analyzed results of standardized protocols for TTI. Giannicola et al.⁵ in a series of 26 patients and Zhang et al.²⁰ analyzing 13 cases reported a mean MEPS of 95 and 85 and mean DASH of 8 and 21, respectively. Nonetheless, despite the satisfactory clinical outcomes reported by the authors, the mean follow-up of the first publication was 31 months and 27 months for the second study, failing to address the sustainability of the results at long-term follow-up.

Despite significant advances in clinical outcomes using standardized approaches for TTI, postoperative complications and reoperations are still of major concerns. After surgical treatment of TTI, Lindenhovius et al.²¹ and Forthman et al.²² reported reoperation rates of 28 and 41%, respectively. In a recent systematic review, Chen et al.³ stated that the most common postoperative complication in TTI’s treatment was HO (12.5% of cases) followed by ulno-humeral arthrosis (11.2% of cases). These authors concluded that despite the high reoperation rate, satisfactory clinical outcomes can be achieved using this protocol. Similarly, we noticed two patients requiring secondary intervention and eight cases showing radiological signs of arthritis at final follow-up. Nonetheless, we did not find correlation between the radiological progression and clinical outcomes according to either MEPS or DASH questionnaire.

The present study has several limitations. First, we had a relatively low number of patients analyzed. However, the very low incidence of such injuries with long-term follow-up may justify the limited number of patients enrolled. Furthermore, to our knowledge this study has the longest follow-up reported in the current literature. Second, it is retrospective in nature, and therefore prone to the same bias and limitations of data collection as other reviews. Third, there was no control group; hence, no conclusions can be made regarding comparisons with other type of treatment methods. Despite these limitations, our findings suggest that using a standardized protocol, satisfactory clinical outcomes at early follow-up are maintained over time. Nevertheless, surgeons should be aware of the possible complications and that radiological changes may be expected at long-term follow-up.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Review and Patient Consent

This retrospective study was carried out with the approval of our Institutional Review Board. All patients consented to collection of data as part of this publication.