Distraction osteogenesis of the humerus: a case of 100% humeral lengthening in a short duration without permanent complications

Abstract

A 15-year-old boy with right humeral shortening underwent distraction osteogenesis with 1 mm/day lengthening. Preoperative Sauvegrain method showed skeletal maturity and lengthening did not account for future growth. Functional loading was encouraged 2 months before end of consolidation phase to hasten consolidation and fixator removal was done in the clinic without local anaesthesia at the patient’s request 1 month earlier. A Sarmiento brace was made and compliance taught to the patient; however, the patient did not wear the brace and resumed body weight exercise weeks after removal of the fixator. Two months after removal of the external fixator, plain radiographs showed good recanalisation of the humeral shaft with no fracture or distortion. However, the left humerus grew and is now 1 cm longer than the right humerus. Nevertheless, the patient was satisfied with the upper limb length discrepancy.

Background

Distraction osteogenesis of the humerus is safe and more commonly done nowadays. The case presents learning points of 100% humeral lengthening on a paediatric patient with low healing index and complication rate.

Case presentation

A 15-year-old boy with history of right proximal humerus fracture managed with closed reduction and percutaneous pinning 12 years prior presented with right upper limb length discrepancy secondary to physeal arrest and relative reduced strength of right upper limb (figure 1). Medical history was otherwise unremarkable. He had a left shoulder abduction of 0–170 degrees and forward flexion of 0–160 degrees. However, right shoulder abduction was 0–140 degrees and forward flexion was 0–120 degrees. Left shoulder internal rotation was until T8. However, that on the right was until L1. He was training to be a powerlifter and the asymmetrical length had affected his progression.

Figure 1.

Preoperative showing humeral length discrepancy.

Investigations

Anteroposterior and lateral plain radiographs of bilateral humeri were taken and showed right humeral length of 10.13 cm and shortening of 10.13 cm (figure 2). The right humerus head was also flattened and non-spherical.

Figure 2.

Preoperative plain radiographs of right humerus.

Treatment

The patient was electively admitted for right humerus osteotomy and distraction osteogenesis. Surgery was performed according to Ilizarov principles.

Prior to surgery, both the patient and his parents were counselled of the risks, complications, challenges and commitment over multiple visits. The lengthening protocol and risks were checked with the patient to ensure understanding and commitment. This was done at least twice prior to surgery and daily postoperatively before discharge. The patient was also reviewed weekly after his discharge to check pin-site hygiene and reinforce fixator care and lengthening protocol. Initial weekly radiographs were also performed to measure adequate distraction speed.

Surgery procedures

The patient was placed supine under general anaesthesia on a regular operating table. Preoperative intravenous first-generation cephalosporin was given. No tourniquet was used. A proximal surface marking was made 5 cm distal to acromion. A distal marking just above metaphysis was made using image guidance. A 2 cm incision was made at the radial groove approximately at the anterolateral arm to locate and protect the radial nerve before fixating the distal threaded rods.

A K-wire was introduced laterally through the distal marking parallel to the elbow axis of rotation. A second pin was introduced laterally at the proximal marking and was ensured to be parallel to the first pin. Image intensifier was used to confirm placement of the wires. Subsequent placement of the unilateral uniplanar external fixator was done with the two pins as references. Three Schanz rods were inserted proximally and another three distally using the monorail fixator as a guide.

A 1 cm longitudinal incision was made in the periosteum mid-shaft. Subperiosteal circumferential dissection was performed using the aid of a periosteal elevator and Homan retractor. A low-speed 2 mm drill was used to create five bicortical transverse holes using a Homan retractor to protect structures at the far cortex. The transverse osteotomy was completed using quarter-inch osteotomes. An image intensifier image was taken to confirm pin placement. The Schanz pins were used to turn the proximal and distal fragments 90 degrees in opposite directions along the longitudinal axis to ensure completion of the osteotomy. The monorail fixator was attached to the Schanz pins and distracted about 1 cm and then used to gently compress the osteotomy site. The pin sites were dressed with povidone-soaked gauze and crepe bandage. The patient was then reviewed 1 week after surgery to confirm callus formation followed by initiation of distraction of 0.25 mm, four times a day.¹

Outcome and follow-up

Postoperatively, the patient experienced wrist drop with sensory deficit over the first dorsal web space. Reassurance was given and wrist range of motion exercises were taught. The patient was discharged from hospital after 3 days following adequate pain control. Weekly follow-ups for the first month were performed to ensure compliance and pin-site care and also to closely monitor the radial nerve palsy in addition to aggressive physiotherapy. The wrist dorsiflexion started improving by the second week and hence the distraction protocol was not altered. There was full radial nerve recovery at the end of 8 weeks of treatment. Subsequently, follow-ups were at monthly intervals.

Lengthening was aimed at the same length of the contralateral humerus. A 3-month lengthening period allowed for 10 cm lengthening, a speed of 1.11 mm/day followed by 6 months of consolidation phase (figures 3–4). One hundred per cent lengthening was eventually achieved and lengthening percentage was calculated by taking lengthening over initial diaphyseal length. Healing index is the number of days of consolidation per centimetre new bone growth and is 24 days/cm for this patient.

Figure 3.

Start of lengthening process.

Figure 4.

End of lengthening process and start of consolidation process.

Following 2 months of lengthening, the patient complained of decreased right elbow range of motion together with occasional clicking. In discussion with the treating surgeon, he was still keen to proceed with lengthening. He was thence taught elbow stretches to be done three times a day. The symptoms had improved after 3 weeks of stretches and was continued on until the end of the lengthening phase with no further recurrence of the elbow stiffness.

Five months into treatment, the patient had a pin-site infection at the most distal and second most distal pin and was managed with povidone-iodine dressing without antibiotics. Infection was subsequently cleared. Lengthening was continued for to a total of 3 months until both upper limbs were equal in length.

After achieving adequate lengthening, consolidation phase then constitutes 6 months, double the duration of the distraction phase. To hasten consolidation, the patient was encouraged to functionally load the arm by doing push-ups 4 months into consolidation. He was supposed to be fixated for another 2 months but requested for removal of fixator 1 month earlier due to personal reasons. The removal was done in the clinic without local anaesthesia at the patient’s request after confirming that the thickness of the regeneration was adequate (figure 5). A Sarmiento functional brace was made and compliance taught. However, the patient did not wear the brace and instead defaulted follow-ups for 2 months. He had resumed bodyweight exercises at 2 weeks following removal against medical advice.

Figure 5.

Before removal of external fixator showing good consolidation.

Two months after removal of the external fixator, plain radiographs showed good recanalisation of the humeral shaft with no fracture or distortion (figure 6).

Figure 6.

Follow-up 6 weeks after removal of external fixator.

One year post-lengthening follow-up, however, revealed that the left humerus grew and is now 1 cm longer than the right humerus (figure 7). Nevertheless, patient was satisfied with the upper limb length discrepancy.

Figure 7.

Follow-up 15 months after surgery showing slightly longer left humerus due to growth.

The functional outcomes were evaluated with two validated questionnaires: the Medical Outcomes Study Short Form-36 Health Survey (SF-36) and the Disabilities of the Arm, Shoulder and Hand (DASH) score. SF-36 consists of multi-item scales measuring eight health domains: physical functioning (PF); role limitations due to physical health (RP); bodily pain (BP); social functioning (SF); vitality, energy or fatigue (VT); general health perceptions (GH); role limitations due to emotional problems (RE); general mental health (MH). The SF-36 scores for the patient were as follows: PF (85), RP (75), BP (45), GH (60), VT (55), SF (50), RE (33.3), MH (40). Preoperative and postoperative DASH scores were 5.8 and 6.7, respectively.

Discussion

A unilateral uniplanar external fixator instead of Ilizarov frame or hybrid fixator was used for its lower complication rate.^2–5 In addition, a monorail external fixator is sufficient to achieve lengthening and avoids the bulkiness of a ring fixator.

Humeral lengthening brings about many complications. Risk of complications increased when the limb is lengthened more than 20%.⁶ However, there have been many studies which showed humeral lengthening of more than 20% with minimal complications. Hosny⁷ analysed 56 humeri lengthening on 46 paediatric patients and reported a mean 55% lengthening. Similarly, other studies reported humeri lengthening ranging from 40% to 60%.^{2 4 8}

Some complications include pin-site infection and peripheral nerve palsy and more severe complications include deep infection or osteomyelitis. It is common to have pin-site infection and peripheral nerve palsy and they can be managed with regular povidone-iodine dressing and physiotherapy, respectively.

Postoperatively, the patient experienced wrist drop with sensory deficit over the first dorsal web space. Decision was made to closely monitor the nerve palsy and to start patient on physiotherapy. The patient was then discharged from hospital after 3 days following adequate pain control. Weekly follow-ups for the first month were performed to ensure compliance and pin-site care and also to closely monitor the radial nerve palsy in addition to aggressive physiotherapy. The wrist dorsiflexion started improving by the second week and hence the distraction protocol was not altered. There was full radial nerve recovery but the end of 8 weeks of treatment. Subsequently, follow-ups were at monthly intervals.

Due to the lack of an established upper limb length predictor like that of Paley multiplier method⁹ for the lower limb, it was difficult to predict the final upper limb length in a growing child. Paley et al ¹⁰ also devised a predictor method for the upper limb; however, it has been rarely used most likely due to the scarcity of paediatric humeral lengthening cases. Sauvegrain method¹¹ showed skeletal maturity and thus we did not expect further growth in the normal humerus.

We note that early functional loading can be instituted to hasten consolidation thereby reducing the length of treatment and risks of complications such as infections of any severity. Patient should be encouraged to slowly resume loading as usual despite early removal of fixator.

Most paediatric humeral lengthening studies reported healing index ranging from average 28 days/cm to 34.5 days/cm and absolute lengthening ranging from average 6 cm to 8.8 cm.3 8 12–16 Particularly of note, Shadi et al found that patients with achondroplasia have a lower healing index of 24.8 days/cm in contrast to 28.6 days/cm for patients without achondroplasia. Our patient achieved lower healing index with higher absolute lengthening than the aforementioned studies, and we attribute our outcome to the reduction of treatment time from encouraging early functional loading.

General mental health, bodily pain and social functioning scores were lower when compared with the general population. Lower MH and SF scores could be attributed to the patient coming from a complex family environment of previous physical abuse episodes. Low BP score was mainly due to pain experienced during loading shortly after removal of external fixator.

Both preoperative and postoperative DASH scores were both below 10 points which supports the fact that the patient did not have much issue with his shorter right humerus, and the main indication for surgery was for his desire to do powerlifting in the future.

Learning points.

• A peripheral nerve palsy does not necessitate cessation of distraction. Resolution of the nerve palsy is the rule as the distraction is gradual and the nerve is in continuity. This allows adequate time for recovery of the nerve.

• Potential growth of contralateral humerus should be accounted for in the lengthening process much as how Paley multiplier method is used for the lower limb length prediction.

• Functional loading hastens bone consolidation and increase in bone density. A lower healing index allows earlier return to activities and decreases the rate of complications. The external fixator can be removed earlier which decreases the risk of pin-site infections or the need for replacement of the external fixator with internal fixation.

• Distraction osteogenesis is high-risk surgery. Indications must be absolute and discussed with both patient and parents prior to surgery and over multiple consultations to allow sufficient time for consideration and assimilation. Compliance has to be checked by patient repetition of risks and lengthening protocol. The authors recommend initial weekly pin-site and radiological checks.