Management of a Humeral Shaft Non-union with Concomitant Rotator Cuff Tear Arthropathy
An 83-year-old male patient with multiple medical comorbidities presents with right arm dysfunction following a mechanical fall 3 months prior. Radiographs reveal a proximal third humeral shaft fracture with varus deformity and minimal early healing, as well as rotator cuff tear arthropathy. What are the treatment options to consider for this patient?
Tyler A. Luthringer, MD; Kevin J. Cronin, MD; Daniel E. Davis, MD, MS; Surena Namdari, MD, MSc
Approximately 85% of humeral shaft fractures heal with conservative management. [1,2] Fracture characteristics that portend a lower union rate absent surgical intervention include proximal third fractures (60% to 76% union rate) and simple spiral or oblique fractures (as opposed to comminuted patterns). [1,3] In addition, recent data have demonstrated ipsilateral glenohumeral arthritis to be an independent risk factor for humeral shaft non-union. 
Patients with humeral shaft non-union often experience continued pain, prolonged disability, and impaired quality of life. Operative treatment of humeral shaft non-union leads to successful healing and improved function in up to 95% of cases; however, delayed definitive management is associated with increased morbidity. 
Rotator cuff tear arthropathy (CTA) is chracterized by progressive glenohumeral arthritis with clearly discernable morphologic features resulting from a massive defect of the rotator cuff. Affected patients typically complain of pain with associated functional limitations, manifest progressive weakness, and loss of both active and passive shoulder motion.  Reverse shoulder arthroplasty (RSA) is a reliable option to relieve pain and restore meaningful shoulder function in patients with CTA. 
In this article, we describe the unique presentation of an 83-year-old, right-hand dominant male with severe right arm dysfunction secondary to a humeral shaft non-union in the setting of ipsilateral CTA. The case highlights the options for, and difficulty in, managing these concomitant pathologies in a medically frail individual who has lost meaningful upper extremity function and independence because of this injury.
An 83-year-old, right-hand dominant male presents to the office with right shoulder pain and inability to use his right arm 3 months after sustaining a proximal third humeral shaft fracture from a low-energy mechanical fall. On acute presentation to an outside hospital, he was treated non-operatively in a sling. He reports pain with any attempt to reach or lift the arm and difficulties with activities of daily living. Prior to the fall, the patient had received 2 cortisone injections for pre-existing shoulder pain and limited overhead motion. He enjoys golfing but has been unable to golf since the injury.
The patient’s medical comorbidities include a prior cerebrovascular accident, non-insulin dependent diabetes mellitus, hypertension, hyperlipidemia, and renal cell carcinoma for which he was scheduled to undergo his third course of treatment not long after presentation to our office.
The decision was made to continue conservative management, which would allow additional time for healing of the proximal humeral shaft fracture while the patient was undergoing medical treatment for renal cell carcinoma. The patient was re-evaluated at 6 months and 9 months from the date of injury, at which time he had gone on to definitive non-union.
Due to persistent pain, significant functional impairment, and poor quality of life, the patient opted for surgical intervention. However, during his preoperative medical assessment, he was found to require additional treatment for renal cell carcinoma and orthopaedic surgery was further delayed.
Fifteen months after the injury, the patient’s medical issues had been appropriately treated and optimized and he was again evaluated for orthopaedic surgery. He complained of the same symptoms: inability to use the right arm, unrelenting pain, and severe sleep disturbance. The patient elected to reschedule surgery with the primary goal of pain relief.
3-months from the date of injury
- Well-groomed male in no distress
- Body mass index of 25.1 kg/m2
- No significant atrophy of the shoulder girdle
- Tender to palpation at the greater tuberosity, biceps groove
- Tender to palpation with gross motion and crepitus of the fracture site
- Palpable prominence of the lateral arm
- Increased lateral arm deformity with deltoid contraction on attempted elevation
- Active motion – forward flexion (FF): shrug, abduction (AB): shrug, external rotation (ER): 0°
- Passive motion – FF: 90°, ER (arm at side): 20°, internal rotation (IR, arm at side): unable to perform
- Rotator cuff motor strength deferred
- Axillary, radial, median, ulnar, musculocutaneous, long thoracic, spinal accessory, and anterior and posterior interosseous nerves intact to motor testing
- Sensation intact to light touch in axillary, radial, median, and ulnar nerve distributions
- 2+ radial pulse
15 months from the date of injury
- Well-nourished male in no distress
- Body mass index: 24.96 kg/m2
- Motion and crepitus at the non-union site with attempted elevation
- Active motion – FF: shrug, AB: shrug, ER: 20° and painful
The patient’s initial post-injury imaging was not available for review. Radiographs on presentation to our institution (3 months from injury) demonstrated right shoulder CTA and a comminuted, proximal third humeral shaft fracture with early callus formation and incomplete healing. There was healing of a large lateral bone fragment to the proximal humerus in a malunited fashion (Figure 1).
Figure 1. Anteroposterior radiographs of the right shoulder in external rotation (left) and internal rotation (right) demonstrate superior migration of the humeral head and a shortened, comminuted, proximal third humeral shaft fracture with varus deformity and a displaced lateral butterfly fragment. There is early callus formation between the proximal and distal fragments medially but no significant healing of the lateral butterfly fragment.
Radiographs 15 months post-injury demonstrated a proximal third humeral shaft non-union (Figure 2). At that time, computed tomography (CT) imaging was obtained for surgical planning (Figures 3-5).
Figure 2. Anteroposterior radiographs of the right shoulder in external rotation (left) and internal rotation (center) and a scapular Y radiograph (left) demonstrate hypertrophic non-union with mobile deformity and resorption between the proximal and distal main fragments. There is apparent bony healing of the lateral butterfly fragment proximally.
Figure 3. Coronal (left) and sagittal (right) CT images demonstrate the chronic fracture of the proximal third humeral shaft with bony non-union and a 1–cm to 3–cm gap between the proximal and distal fragments, as well as healed malunion of the lateral butterfly fragment with bridging callus to the proximal fragment and superior migration of the humeral head with remodeling of the overlying acromion.
Figure 4. Axial CT imaging demonstrates neutral glenoid version.
Figure 5. Three-dimensional CT reconstruction images of the humeral non–union and deformity.
- Proximal third humeral shaft hypertrophic non-union with ipsilateral rotator cuff tear arthropathy
Non-operative management of isolated, closed humeral shaft fractures is generally considered first-line treatment due to high rates of successful healing.
- Approximately 85% union rate [1,2,8]
- Avoids risks of complications associated with surgery and anesthesia
- May be preferable in low-demand, medically frail patients
- Offers comparable functional outcomes to operative management of humeral shaft fractures upon successful healing 
- Clinical motion at the fracture site 6 weeks after injury has been shown to be a predictive risk factor for progression to non-union 
- Prolonged period of morbidity related to fracture brace/immobilization
- Higher rate of non-union and reintervention compared with fractures treated primarily with surgery [3,8,9]
Humeral Shaft Non-union Repair with Open Reduction Internal Fixation (ORIF)
Repair of humeral shaft fracture non-union yields successful outcomes of bony union in the majority of cases.  Stabilization can be achieved using plate and screw fixation, intramedullary nailing, or external fixation. Autologous bone graft and/or a variety of allograft options may be used as biologic adjuncts to promote healing. Bone grafting is recommended in patients with atrophic or oligotrophic non-union or in case of a cortical or segmental defect. The cost of donor site morbidity must be weighted against the potential added benefit when using autograft compared with allograft.
- Offers a relatively less invasive and less surgically complex intervention compared with non-union repair with simultaneous RSA
- Union rates of 98% and 95% when plate and screw fixation is used with and without autologous bone graft, respectively 
- Intramedullary nailing or external fixation are alternative options that may afford adequate stability by means of less extensive dissection in certain scenarios
- Fails to address coexisting shoulder pathology in our patient and, therefore, would portend unpredictable function and incomplete pain relief
- Lower union rates demonstrated with intramedullary nailing than with plate fixation 
- Highest complication rate, longest time to union, and added morbidity associated with external fixation method 
Humeral Shaft Non-union Repair with ORIF and RSA
Due to the rarity of the presented indication, techniques to manage humeral shaft non-union in the setting of CTA are not commonly described or reported. Treatment methodology may be drawn from surgeon experience and the literature on managing humeral non-union, periprosthetic fracture, and revision shoulder arthroplasty.
- Addresses both sources of potential pain and dysfunction for this patient, optimizing the chance of a favorable functional outcome
- Use of a long humeral stem to bypass the non-union site offers the additional mode of intramedullary stabilization compared with non-union repair with plate fixation alone
- More extensive operation with longer operative time, higher blood loss, and greater risk of surgical complications
- Greater challenge in achieving hardware fixation in the humerus in the presence of a long cemented humeral stem
- Successful outcome predicated on simultaneous healing of humeral non-union repair and stable fixation of RSA
After a thorough discussion of these options, the patient verbalized his understanding and agreed to undergo the recommended treatment of humeral shaft non-union repair with ORIF and RSA.
Approach and Humeral Head Osteotomy
- The patient was taken to the operating room and positioned in a well-padded beach chair position. General endotracheal anesthesia was induced.
- A standard anterior deltopectoral incision was made. The deltoid and cephalic vein were taken laterally; the pectoralis major was taken inferomedially.
- Subdeltoid and subacromial adhesions were bluntly released.
- The clavipectoral fascia was incised and the axillary nerve was palpated and protected.
- The inferior humeral circumflex vessels were suture ligated and the subscapularis tendon was tagged with a non-absorbable suture.
- The biceps tendon was traced proximally through the rotator interval and released from the supraglenoid tubercle.
- The subscapularis was then peeled from its insertion on the lesser tuberosity to the 6 o’clock position.
- The shoulder was dislocated anteriorly; no remaining posterosuperior rotator cuff was noted.
- Marginal osteophytes were resected from the anatomic neck and a 135° humeral head osteotomy was made with approximately 20°of retroversion.
Glenoid Exposure and Instrumentation
- Deep glenoid retractors were placed.
- The superior glenohumeral, middle glenohumeral, coracohumeral, and inferior glenohumeral ligaments were released while protecting the axillary nerve.
- Remnant labral tissue and the biceps stump were excised.
- The glenoid surface was debrided of any remaining cartilage prior to instrumentation for reaming.
- A centrally threaded glenoid baseplate was placed followed by 4 peripheral locking screws, achieving excellent fixation.
- Peripheral reaming was performed prior to impaction of a size 36 mm glenosphere with 6 mm of lateral offset and placement of a central screw.
- Attention was turned to the humeral shaft.
Humeral Shaft Exposure, Non–union Repair, Fixation, and Instrumentation
- The incision was extended distally into an anterolateral approach to the humerus.
- The pectoralis major tendon was taken down from its humeral insertion and tagged for later repair. The anterior deltoid was also tagged and reflected posterolaterally for later repair (Figure 6).
Figure 6. Intraoperative photograph shows the released and tagged pectoralis major tendon medially, reflected and tagged anterior deltoid laterally, and reduction of the diaphysis at the medial cortex with provisional fixation using a 2.7 limited contact dynamic compression (LC-DC) plate anteriorly.
- The musculocutaneous nerve was identified and protected and the biceps was retracted medially after incision of the biceps sheath.
- The brachialis was split midline to expose the underlying humeral shaft.
- The radial nerve was identified proximally and medially in front of the latissimus and was then traced around the humerus to ensure that it was protected and free from the fracture non-union site.
- The fracture was noted to have malunited proximally and hypertrophically non-united distally. The fracture fragments were mobilized and extensively debrided of all fibrous non-union tissue.
- Fracture edges were burred and sawed to healthy bleeding bone.
- The humeral canal was initially opened and reamed proximally and distally through the fracture site.
- A reamer was then placed down the canal from proximal to distal to align and reduce the medial cortex. Provisional fixation was achieved with a 2.7 LC-DC plate, 2 unicortical screws proximally, and 3 unicortical screws distally (see Figure 6, above).
- The humeral canal was reamed and trialed.
- Trial components were removed and the canal was prepared for cementation. A cement restrictor was placed distally.
- Cement was pressurized proximally and distally with care to ensure no extravasation out of the fracture non-union site.
- A size 6 long stem (170 mm) humeral component was inserted and cement was allowed to cure.
- Attention was then turned to the proximal malunion. The lateral butterfly fragment was osteotomized, realigned, and fixed with 2 lag screws (2.7 mm).
- Recombinant human bone morphogenetic protein-2 (rhBMP-2) allograft was placed at the fracture non-union site.
- A 13-hole proximal humerus locking plate was applied anterolaterally and affixed with 4 bicortical non-locking screws in the distal segment and 6 bicortical non-locking screws proximally.
- Intraoperative fluoroscopy confirmed fracture reduction and hardware position (Figure 7).
Figure 7. Intraoperative fluoroscopic images show the final position of all implants and hardware.
Trial Reduction, Polyethylene Insertion, and Wound Closure
- The shoulder was reduced and trialed and a +4 semi-constrained humeral liner was selected.
- The wound was irrigated, the final polyethylene liner was inserted, and the shoulder was again reduced and found to be stable.
- Non-absorbable #5 sutures were used to repair the previously tagged anterior deltoid and pectoralis major through the plate, to tenodese the biceps tendon to the plate anteriorly, and to repair the subscapularis (Figure 8).
Figure 8. Intraoperative photograph demonstrates preparation of the repair of the pectoralis major tendon, anterior deltoid, and the biceps tenodesis through the plate. Supscapularis repair was performed with 3 #5 non-absorbable sutures passed through bone at the bicipital groove.
- The shoulder was brought through the range of motion and was again noted to be stable.
- The wound was irrigated a second time and closed in a layered fashion.
The patient was placed in a sling with an abduction pillow, admitted to a hospitalist service, and moved to the telemetry unit for monitoring postoperatively. He remained neurovascularly intact with relatively well-controlled pain. Significant events of the hospital course included acute kidney injury, transfusion for acute blood loss anemia on POD2, and new-onset paroxysmal atrial fibrillation on POD3. The patient stabilized and was medically cleared for discharge home with home health services on POD4.
On POD11, the patient experienced new-onset dyspnea, which led him to present to an outside facility. He ultimately underwent a triple coronary artery bypass procedure and pacemaker and defibrillator placement prior to his first postoperative visit.
4 Weeks Postoperatively
The patient was seen in the office for the first time at 4 weeks post-surgery. The incision was well-healed without any signs of infection. He had passive FF to 80° and passive ER to 30° (arm at the side). Radiographs showed no signs of hardware loosening or failure.
The patient was instructed to avoid any pushing, pulling, or lifting activity and to continue independent passive motion exercises at home.
3 Months Postoperatively
The patient reported improvement in pain and motion and continued compliance with activity restrictions and his home exercise program.
On examination, the patient had active FF to 60°, AB to 30°, ER to 35°, and IR to S1. His passive motion was to 90° FF and 40° ER.
A formal home physical therapy program was initiated, with continuation of lifting restrictions.
6 Months Postoperatively
The patient continued to progress in his recovery without complication. On examination, active motion remained unchanged with the exception of ER to 10°. Passive FF was to 130° and ER was to 45° Radiographs demonstrated interval consolidation of the non-union site and no signs of hardware loosening or failure (Figure 9).
Figure 9. Anteroposterior radiographs of the right shoulder in external rotation (left) and internal rotation (center) and a scapular Y radiograph (left) obtained 6 months postoperatively demonstrate a well-positioned, long stem RSA with stable cement mantle, consolidation of the non-union site, and no signs of hardware loosening or failure.
The patient was allowed to advance his activity as tolerated but was cautioned to avoid heavy and repetitive lifting that could impact implant longevity.
The present case describes the difficult management of a proximal third humeral shaft non-union with ipsilateral CTA in an older patient with a complex medical history. A successful surgical outcome was achieved by addressing concomitant pathologies with a long stem, cemented RSA and humeral non-union with ORIF and allograft. Perioperative medical co-management is crucial for the safety and care of high-risk patients.
Tyler A. Luthringer, MD, is a shoulder and elbow surgery fellow at The Rothman Institute, Philadelphia, Pennsylvania.
Kevin J. Cronin, MD, is an orthopaedic surgeon with the Florida Orthopaedic Institute in Tampa, specializing in the treatment of shoulder and elbow conditions. He is a former shoulder and elbow surgery fellow at The Rothman Institute, Philadelphia, Pennsylvania.
Daniel E. Davis, MD, MS, is an orthopaedic surgeon with The Rothman Institute, Philadelphia, Pennsylvania, specializing in the treatment of shoulder and elbow conditions. He is also the Shoulder Section Editor for Rothman Institute Grand Rounds on ICJR.net.
Surena Namdari, MD, MSc, is a an orthopaedic surgeon with The Rothman Institute, Philadelphia, Pennsylvania, specializing in the treatment of shoulder and elbow conditions. He the Shoulder and Elbow Fellowship Director and the Co-Director of Shoulder and Elbow Research.
Disclosures: Dr. Luthringer has no disclosures relevant to this article. Dr. Cronin has disclosed that he is a paid presenter or speaker for DJO. Dr. Davis has disclosed that he is a consultant for Arthrex, Inc. and that owns stock/stock options in Catalyst OrthoScience. Dr. Namdari has disclosed that he receives royalties and research support from and is a paid presenter/speaker for DJO, that he receives royalties from Miami Device Solutions/Biederman Motech; that he is a paid consultant and paid presenter/speaker for Miami Device Solutions; that he receives royalties and owns stock/stock options in Aevumed; that he owns stock/stock options in Coracoid Solutions, Force Therapeutics, HealthExl, Orthophor, Parvizi Surgical Innovations, MD Live, MD Valuate, Rothman Institute, RubiconMD, and Tangen; and that he receives research support from Arthrex, Inc., DePuy – A Johnson & Johnson Company, Integra, Roche, Wright Medical Technology, Inc., and Zimmer.
- Ali E, Griffiths D, Obi N, Tytherleigh-Strong G, Van Rensburg L. Nonoperative treatment of humeral shaft fractures revisited. J Shoulder Elb Surg 2015;24:210–4. doi:10.1016/j.jse.2014.05.009.
- Driesman AS, Fisher N, Karia R, Konda S, Egol KA. Fracture site mobility at 6 weeks after humeral shaft fracture predicts nonunion without surgery. J Orthop Trauma 2017;31:657–62. doi:10.1097/BOT.0000000000000960.
- Harkin FE, Large RJ. Humeral shaft fractures: union outcomes in a large cohort. J Shoulder Elb Surg 2017;26:1881–8. doi:10.1016/j.jse.2017.07.001.
- Oliver WM, Searle HKC, Ng ZH, et al. Factors associated with humeral shaft nonunion. J Shoulder Elb Surg 2021. doi:10.1016/j.jse.2021.01.029.
- Peters RM, Claessen FMAP, Doornberg JN, Kolovich GP, Diercks RL, Van Den Bekerom MPJ. Union rate after operative treatment of humeral shaft nonunion – A systematic review. Injury 2015;46:2314–24. doi:10.1016/j.injury.2015.09.041.
- Neer CS, Craig E V, Fukuda H. Cuff-tear arthropathy. J Bone Joint Surg Am 1983;65:1232–44.
- Coscia AC, Matar RN, Espinal EE, Shah NS, Grawe BM. Does preoperative diagnosis impact patient outcomes following reverse total shoulder arthroplasty? A systematic review n.d. doi:10.1016/j.jse.2020.10.003.
- Van De Wall BJM, Ochen Y, Beeres FJP, et al. Conservative vs. operative treatment for humeral shaft fractures: a meta-analysis and systematic review of randomized clinical trials and observational studies. J Shoulder Elb Surg 2020;29:1493–504. doi:10.1016/j.jse.2020.01.072.
- Lode I, Nordviste V, Erichsen JL, Schmal H, Viberg B. Operative versus nonoperative treatment of humeral shaft fractures: a systematic review and meta-analysis. J Shoulder Elb Surg 2020;29:2495–504. doi:10.1016/j.jse.2020.05.030.