The “Bag of Bones” Treatment of Comminuted Intra-articular Distal Humerus Fractures in the Elderly

Abstract

Background:

Extensively comminuted intra-articular distal humerus fractures in the elderly present a challenging therapeutic dilemma. The purpose of this study was to investigate the results of nonoperative treatment of these fractures in a select subset of patients.

Methods:

Patients treated with nonoperative management for a comminuted intra-articular distal humerus fracture between 2007 and 2018 were reviewed. Patients were administered 3 elbow-specific functional outcomes instruments.

Results:

A total of 8 patients (2 men, 6 women) were treated with brief immobilization followed by early range of motion. All had fractures with extensive comminution of the articular surface such that open reduction and internal fixation was not feasible. Average age was 70 years. At an average of 33 months postinjury, average flexion was 124°, and extension was −27°, with full forearm rotation. No patients required pain medications at the latest follow-up. At 33 months of follow-up, the average Mayo Elbow Performance Score was 92/100 (100 optimal), Oxford Elbow Score was 43/48 (48 optimal), and Quick Disabilities of the Arm, Shoulder, and Hand Score was 10/100 (0 optimal). All patients were satisfied with the nonsurgical treatment that they received.

Conclusions:

Nonoperative treatment for comminuted intra-articular distal humerus fractures results in acceptable functional outcome in elderly patients and should be considered when the fracture is not amenable to internal fixation and in lower-demand patients with higher surgical risk.

Introduction

Nonoperative treatment of distal humerus fractures, later termed the “bag of bones” approach, was first described over 75 years ago. ¹ This approach consists of early mobilization for patients who sustained a distal humerus fractures. With advances in surgical techniques, operative management of these fractures has shown improved outcomes.^2,3 Consequently, the bag of bones approach is primarily reserved for patients with distal humerus fractures who are too medically unwell to undergo a surgical procedure. ⁴

The contemporary approach to treatment of comminuted intra-articular distal humerus fractures in the elderly population has focused on total elbow arthroplasty (TEA) and open reduction and internal fixation (ORIF).^{5 -8} Open reduction and internal fixation has shown promising results; however, there is often osteopenia, low articular fractures, and extensive comminution of the articular surface that makes ORIF more challenging in this patient population. This has led some authors to advocate TEA as the treatment of choice in this patient population.^{5,6,9 -12} Complications are common with both ORIF and TEA.^{13 -16} Patients treated with arthroplasty have a lifetime lifting restriction, and arthroplasty has several potential serious complications including loosening, wound breakdown, infection, periprosthetic fracture, and implant failure.^13,17,18 The potential disadvantages of operative management have led to some interest in nonoperative management of distal humerus in intraarticular fractures in elderly.^4,19

The purpose of this study was to investigate the functional outcomes of “bag of bones” approach to treatment of elderly patients with severely comminuted intraarticular distal humerus fractures not amenable to internal fixation.

Materials and Methods

This investigation was conducted as a retrospective review of patients treated with nonoperative management of an extensively comminuted intraarticular distal humerus fracture. The study was approved by the institutional review board, and consent was obtained from patients to complete functional outcomes instruments. All patients in this series presented to and were treated by the senior author between the years 2007 and 2018. Patients were identified by review of the senior author’s records. Inclusion criteria consisted of age 60 years or older at time of injury, presence of a comminuted intraarticular distal humerus fracture (Müller AO Classification types 13B and 13C), and initial definitive nonoperative management. Exclusion criteria consisted of age less than 60 years at time of injury and treatment with operative management.

Nonoperative treatment consisted of a brief period of immobilization which varied from 5 days to 2 weeks, in a sling or splint for comfort, followed by early range of motion (ROM) of the elbow and forearm. Each patient participated in a course of formal occupational therapy to work on ROM and strengthening. Range of motion exercises were performed in a supervised as well as unsupervised manner, the latter in the form of a home program.

Medical records were reviewed for age, gender, laterality, mechanism of injury, hand dominance, medical comorbidities, and follow-up with respect to ROM and functional limitations. Patients were administered 3 elbow-specific functional outcomes instruments: the Mayo Elbow Performance Score (MEPS), ²⁰ the Oxford Elbow Score, ²¹ and the Quick Disabilities of the Arm, Shoulder, and Hand (QuickDASH) Score. ²²

Results

There were 8 patients in our series (Table 1), 2 men and 6 women. The mean age at time of injury was 70 years (range: 63-82 years). All patients sustained a comminuted intraarticular distal humerus fracture after a mechanical fall from standing height (Figures 1 and 2). Fracture characteristics such as extensive comminution, low articular fractures, and poor bone quality guided the senior author in choosing nonoperative management. Four patients injured their dominant arm, and 4 patients injured their nondominant arm (Table 1). Medical comorbidities of patients in our series are listed in Table 2.

Table 1.

Patient Demographics and Outcomes.

Patient	Age, y	Sex	Side	Hand dominance	Extension (°)	Flexion (°)	Pronation	Supination	F/U	Pain	Mayo Elbow Performance Score	Oxford Elbow Score	Quick DASH score
1	82	M	R	R	30	100	Full	Full	37	None	85	45	2.3
2	73	F	L	L	10	140	Full	Full	57	Occasional clicking	100	46	2.3
3	73	F	R	L	46	135	Full	Full	24	None	85	32	0
4	71	M	L	R	20	135	Full	Full	39	Occasional	100	48	40.9
5	64	F	L	R	12	132	Full	60	49	Occasional clicking	85	43	4.5
6	63	F	L	R	30	130	Full	Full	7	Occasional	NA a	NA a	NA a
7	70	F	R	R	35	125	60	Full	44	Intermittent soreness	95	47	12
8	65	F	R	R	35	95	Full	Full	5	Mild soreness	NA a	NA a	NA a
Average	70				27	124			33		92	43	10

Note. F/U = average follow-up in months; QuickDASH = Quick Disabilities of the Arm, Shoulder, and Hand; NA = not applicable; M = male; F = female; L = left; R = right.

^aPatient 6 deceased from metastatic cancer, and patient 8 has ipsilateral resection arthroplasty of the shoulder after multiple infections.

Figure 1.

Posteroanterior and lateral radiographs.

Note. (a) Represents a comminuted intraarticular distal humerus fracture with comminution of the lateral column and shear component of capitellum and trochlea. (b) Representative commuted tomography images (coronal and 3-dimensional formatting) show low intra-articular distal humerus fracture with shear of the capitello-trochlear fragment and comminution of lateral epicondyle and lateral column.

Figure 2.

Posteroanterior and lateral radiographs at presentation (a) with extensively comminuted intra-articular distal humerus fracture. (b) Fracture union at 6 months.

Table 2.

Medical Comorbidities, Decision-Making, and Management.

Patient	Medical comorbidities	Decision-making a	Treatment	X-ray F/U	X-ray	Dynamic splinting
1	Atrial fibrillation on Coumadin, hypercholesterolemia, hypertension	Extensive comminution, age	Long arm cast, ROM started at 3 wk	37	Elbow arthritis, heterotopic ossification, healed at 3 mo	At 3 mo for 45-100 motion
2	Peripheral neuropathy, hypercholesterolemia, rheumatoid arthritis, osteopenia	Extensive comminution	Long arm splint, ROM started at 3 wk	57	Mild arthritis, healed at 4 mo	No
3	Hypertension, asthma, Barrett esophagus, Parkinson disease, right reverse total shoulder arthroplasty	Fracture comminution, obesity, ipsilateral shoulder prosthesis	Long arm splint, ROM started at 2 wk	12	Mild arthritis, healed at 6 mo	No
4	Hypertension, benign prostatic hypertrophy, colectomy, fixation of tibial plateau, nasal polyp excision	Extensive comminution	Sling, ROM started at 2 wk	3	Healing fracture	No
5	Prior left radial head excision and lateral ligament repair, diabetes mellitus II, osteoporosis, obesity, depression, ameloblastoma of mandible, hypertension, hypercholesterolemia	Extensive comminution, morbid obesity	Long arm splint, ROM started at 3 wk	49	Mild arthritis, heterotopic ossification, healed at 3 mo	No
6	Frequent falls, dementia, depression, seizures, breast cancer, rectal cancer, hypertension, hypothyroidism, osteoporosis	Cognitive dysfunction, medical comorbidities, fracture comminution	Sling, patient noncompliant and started ROM on own	7	Heterotopic ossification, preserved joint space, healed at 4 mo	No
7	Anxiety, diabetes mellitus II	Extensive comminution	Long arm splint, ROM started at 2 wk	9	Preserved joint space, healed at 5 mo	Yes
8	Morbid obesity, multiple shoulder surgeries for infected total arthroplasty and finally right excision shoulder arthroplasty	Infected humerus, morbid obesity, resection arthroplasty of shoulder	Sling, ROM started at 2 wk	5	Preserved joint space, healed at 5 mo	No

Note. F/U = final radiographic follow-up in months; ROM = range of motion.

^aAll patients had extensive comminution not amenable to internal fixation.

At most recent follow-up at an average of 33 months (range: 5-57 months), average flexion was 124° (range: 95°-140°) and average extension was −27° (range: −10° to −46°). Excellent pronation and supination were achieved for all the patients (Table 1, Figures 3 and 4). No patients required pain medications at latest office follow-up, and all patients returned back to full preinjury activities. Two patients experienced ulnar paresthesias that resolved within 3 months after injury. Fractures healed at 4 to 6 months for most of the patients (Table 2, Figure 2). For one patient, there was only a 3-month radiographic follow-up and the fracture was still healing (Table 2).

Figure 3.

Clinical images of the patient in Figure 1 after bag of bones treatment. Slight limitation in (a) extension and (b) flexion is seen. Practically, (c) full supination and (d) pronation are achieved.

Figure 4.

Clinical images of the patient in Figure 2 after bag of bones treatment. Limitation in (a) extension, (b) flexion, and (c) pronation. (d) Full supination is achieved.

Two patients were treated with static progressive splinting (Joint Active Systems, Inc, Effingham, Illinois) at 3 to 4 months postinjury to improve elbow flexion-extension arc.

Outcomes instruments were administered to 6 of the 8 patients (Table 1). Average MEPS was 92 out of 100 (range: 85-100). Mean Oxford Elbow Score was 43 out of 48 (range: 32-48). Average QuickDASH score was 10 (range: 0-40.9), with an optimal score being 0. All patients were satisfied with the nonsurgical treatment that they received.

Discussion

Comminuted intra-articular distal humerus fractures in elderly patients are often challenging for the treating orthopedic surgeon. Advances in internal fixation have made it the treatment of choice in younger people, and there have been reports of successful outcomes of ORIF in the elderly population as well. In 1985, Jupiter et al ²³ reported 62% excellent results and 31% good results in 13 patients aged 65 and older treated with ORIF for a distal humerus fracture. Four out of 13 of these patients experienced a complication. John et al ²⁴ studied 49 patients of average age 80 years at mean 1.5-year follow-up. This series found 31% of patients with excellent results and 49% with good results. Complications consisted of 3 plate fractures and 1 nonunion. In 1997, Pereles et al ⁸ reported their experience with 12 patients of mean 71 years old treated with ORIF for a distal humerus fracture. Three patients (25%) had excellent MEPS, whereas the remaining 75% had good MEPS. Complications occurred in two patients consisting of hardware loosening, with one of these patients requiring hardware removal. Liu and colleagues ²⁵ found 78% of patients had an excellent MEPS at mean 2-year follow-up for elderly patients treated with ORIF. Although ORIF may yield good results in experienced hands, there are significant risks inherent in this surgical procedure. A series of 19 elderly patients of mean 72 years old found that 79% of patients had an excellent MEPS at mean 8-year follow-up; however, there was a 10% incidence of early complication. ²⁶ Korner et al ²⁷ reported on 45 patients aged 60 or older treated with ORIF for a distal humerus fracture. These authors found a significant complication rate of 29%, most due to implant failure and/or screw loosening, and 7 of these patients (16% of the series) required revision surgery. Only 20% scored an excellent MEPS, although an additional 38% had a good MEPS. Low articular fractures, significant comminution of trochlear and/or capitellar surface, and poor bone quality among other factors make ORIF particularly challenging for elderly patients.^5,6

Total elbow arthroplasty has gained popularity especially for comminuted intra-articular distal humerus fractures which are difficult to fix with ORIF. In 2004, Kamineni and Morrey ²⁸ reported their experience in 49 patients of mean age 69 years treated with primary TEA. At average 7-year follow-up, mean MEPS was 93 and mean ROM was 107°, although there were 14 complications (29%) and 10 revisions (23%). Ray et al ¹⁰ reported a case series of 7 patients of mean 82 years of age at average follow-up of 3 years. In this series, patients scored a mean MEPS of 92 and had a mean arc of motion of 110°, although there were 2 complications and no revisions. Another study in 2002 found a mean MEPS of 93 and mean ROM of 101° in 19 patients at mean follow-up of 3 years with the development of 2 complications in this cohort. ²⁹ Gambirasio et al ¹² reported a mean MEPS of 94 and mean ROM of 102° in 10 patients at mean follow-up of 1.5 years with no complications and no revisions.

Several studies have compared ORIF and TEA for the treatment of comminuted intra-articular distal humerus fractures in an elderly population. A 2003 investigation by Frankle et al ⁵ of women aged 65 and older found excellent outcomes in 90% treated with TEA versus only 33% of patients treated with ORIF. The authors thus recommended TEA for treatment of this fracture in a population of women aged 65 and older. McKee et al ⁶ who performed a multi-center, prospective, randomized, controlled study comparing ORIF and TEA in elderly patients found improved MEPS at short term and 2-year time points in TEA patients. These authors found that patients treated with TEA had a lower reoperation rate as compared with ORIF. They recommended TEA for patients with distal humerus fractures not amenable to internal fixation. Although TEA has gained increasing popularity as a primary treatment for elderly patients with a comminuted intra-articular distal humerus fractures, it does have risks.^16,30 A recent study ³⁰ on 10-year follow-up of TEA for distal humerus fractures reported a mean arc of motion of 99° and MEPS of 90. There was an 11% incidence of deep infection and additional 11% with periprosthetic fractures. The authors recommended selected use of TEA for lesser active, older patients and cautioned regarding some major complications associated with TEA. One study found 41% rate of reoperation after TEA at a mean follow-up of 6.1 years. ¹⁶ The authors found higher chances of reoperation and implant revision in trauma-related TEA. Moreover, TEA necessitates a lifetime lifting restriction, and patients may not find this restriction acceptable.

Nonoperative treatment for distal humerus fractures was described over 75 years ago by Eastwood. ¹ Eastwood described a technique, first proposed by Hugh Owen Thomas and later popularized by Sir Robert Jones, that consisted of manipulation under anesthesia followed by brief immobilization and then early ROM. This report described 14 patients of mostly younger age, the youngest patient being 16 years old. Eastwood ¹ noted significant pain relief but a residual lack of elbow extension with the use of this method. In 1953, Evans ³¹ coined the term “bag of bones” to describe the technique of nonoperative management with initiation of early elbow ROM. He noted that this technique is best used for elderly patients and should not be used for younger patients due to the loss of power and risks of instability and posttraumatic arthritis. Around the same time, Knight ³² noted that a nonoperative approach may provide a “surprisingly good result” in elderly patients. With the popularity of surgical options in recent times, nonoperative management is primarily reserved for medically unwell patients with high surgical risk. There have been recent reports investigating nonoperative management of distal humerus fractures in the elderly; however, all these studies group extraarticular as well as intra-articular fractures together in their results.^4,19,33,34 Aitken et al ³³ reported only modest results with conservative management of distal humerus fractures. They had a union rate of 53%, Oxford Elbow Score of 30, and a DASH score of 38. They reported on 40 patients of which only 14 had complete intra-articular fractures. They reported outcomes on 20 out of 40 patients; however, it is not clear how many of those 20 patients had intra-articular fractures. Desloges et al ⁴ reported on 19 medically unwell, lower demand elderly patients. Eighty-four percent of these patients had intra-articular distal humerus fractures. Their nonoperative protocol consisted of a long arm cast for an average of 5 weeks and supervised therapy after signs of radiographic healing. At a mean follow-up of 27 months, they reported an average 100° arc of motion and MEPS of 90. Batten et al ¹⁹ reported on 19 patients with comminuted distal humerus fractures. Fourteen out of these were treated nonoperatively using sling for comfort and early ROM. Only half of the patients treated nonoperatively had intra-articular fractures. Five patients were treated directly with a TEA, 80% of which had intra-articular fracture. Range of motion was not reported in this study; however, they showed good Oxford Elbow Score and Visual Analog Score for pain for both the groups. At average follow-up of 55 months, none of the patients treated nonoperatively converted to TEA. Sudah et al ³⁴ performed a systematic review showing acceptable functional outcomes and low rates of conversion to delayed surgery after nonoperative treatment of the distal humerus fractures in the elderly.

This present study looks at functional outcomes of bag of bones management of comminuted intra-articular distal humerus fractures not amenable to internal fixation. Average arc of motion achieved was 97° with excellent functional outcomes (Table 1). Our results are comparable to those reported for surgical management of these injuries in contemporary literature (Table 3). Slightly better ROM has been reported for TEA; however, functional outcomes are not superior and there is a significant risk of complications and lifetime lifting restrictions.

Table 3.

Outcomes of Our Cohort of Patients at Average 33 Months Compared With Other Recent Studies on Treatment of Comminuted Intra-articular Distal Humerus Fractures in Elderly.

Outcomes Comparison	Range of motion (flexion/extension)	Mayo Elbow Performance Score
Present study	97°	92
Egol et al 7
ORIF	98°	85
TEA	92°	79
McKee et al 6
ORIF	95°	73
TEA	107°	86
Frankle et al 5
ORIF	100°	81
TEA	113°	95
Barco et al 30
TEA	99°	90

Note. ORIF = open reduction and internal fixation; TEA = total elbow arthroplasty.

Limitations of our study include the retrospective design and the small number of patients. It is possible that with a larger series, there might be some unfavorable results; however, TEA is always an option for patients with unsatisfactory results. In addition, longer term follow-up may influence outcomes; however, all patients in our series had highly favorable functional outcomes scores at most recent follow-up. These results are applicable only to elderly patients who might have lower functional demands and might not apply to younger patients with higher demands. Patient selection for nonoperative management is important. The decision to treat nonoperatively was made by the senior author based on presence of extensive comminution of the articular surface. In addition, presence of multiple medical comorbidities, morbid obesity, and ipsilateral shoulder replacement or infected humerus in some patients also contributed to decision-making (Table 2).

As demonstrated in our case series, the bag of bones approach is safe and efficacious, with minimal morbidity and few potential complications. Use of this treatment does not preclude future TEA. ³⁵ In addition, if TEA is performed for an arthritic and painful elbow in the future, it might be less challenging compared with TEA after prior ORIF. ¹⁵

In conclusion, elderly patients with a comminuted intra-articular distal humerus fracture can have good functional outcomes with nonsurgical management. As with ORIF or TEA, there is a residual decreased flexion-extension arc after nonoperative treatment; however, this does not appear to significantly impact functional outcome scores in elderly patients. Nonoperative treatment does not preclude subsequent TEA, should the elbow later become painful. The bag of bones approach continues to be relevant and should be considered in older patients with comminuted intra-articular distal humerus fracture not amenable to internal fixation and in medically unwell patients with high surgical risk.